On-Line Decision Support Tools for
Managing Emerging Pathogen Risks in Wastewater
sponsored by the Water Environment Research Foundation
The 2014 Ebola epidemic highlighted a critical need to evaluate risks to wastewater and sewer workers who could be exposed to Ebola and other pathogens discharged in hospital sewer systems. Current practice in the United States allows hospitals to flush liquid wastes from infected patients directly into the toilet, shower, or sink without disinfection. During the Ebola epidemic, some utilities connected to hospitals treating Ebola patients reported worker absences due to fears of infection. While the likelihood of an Ebola epidemic affecting the United States may be small, experiences during the recent outbreak highlighted the need to re-examine hospital practices for disposal of patient waste into sanitary sewers without treatment. Indeed, prior studies suggest that wastewater and sewer workers experience serious infectious diseases such as hepatitis A at much higher rates than the general population, further underlining the need to consider the adequacy of protections for this essential workforce.
In order to answer wastewater utilities’ concerns about hospital handling of Ebola liquid waste, the Water Environment Research Foundation funded a study by the Water Institute of the University of North Carolina (UNC) at Chapel Hill to develop a protocol for managing liquid waste from Ebola patients prior to sewer disposal and to quantify risks under alternative waste management protocols.
This report presents the final results of the study. Importantly, although the study focuses on risks from Ebola virus, the results and recommendations can inform planning for future outbreaks involving other virulent pathogens.
The study’s objectives were to:
- review existing protocols for the disposal of liquid waste of Ebola patients via sanitary sewers;
- develop a protocol for Ebola patient liquid waste disposal reflecting a consensus of opinions among infectious disease specialists, environmental microbiologists, public health agency personnel, and wastewater utilities; and
- quantify the probability that a wastewater or sewer worker stationed at various points along the sewer network could become ill as a result of Ebola patient waste discharge into the sanitary sewer under different waste management scenarios.
Review of Existing Protocols
Protocols for the management of liquid waste from infected Ebola patients published during the peak of the recent Ebola outbreak (June 2014 to February 2015) were identified through consultation with the World Health Organization (WHO), Centers for Disease Control and Prevention (CDC), U.S. Occupational Safety and Health Administration (OSHA), and web searches. In addition, other relevant CDC and OSHA standards were examined. Protocols were characterized according to their processes for handling patient liquid wastes prior to flushing down the toilet, shower drain, or sink drain.
Consensus Protocol Development
To develop a consensus method for managing Ebola patient liquid waste prior to sewer disposal, the UNC Water Institute held a one-day workshop (May 21, 2015) during the 2015 UNC Water Microbiology Conference. Thirty-three participants representing wastewater utilities (n=4), relevant federal agencies (n=7), state health agencies (n=4), industries (n=4), and scientists with relevant expertise (n=14) were invited to attend. In advance of the workshop, participants received a white paper outlining the requirements of existing protocols. During the workshop, participants divided into groups to discuss critical concerns and then reconvened to reach consensus on how Ebola patient liquid wastes should be managed prior to sewer disposal. The UNC team then drafted a schematic of the proposed waste management protocol along with a description of the underlying rationale. The draft was circulated to workshop participants for comments, and the schematic was revised and redistributed to participants for final approval.
Risks to Wastewater and Sewer Workers
To characterize risks to wastewater and sewer workers employed in collection systems receiving discharges from hospitals treating Ebola patients, the UNC team constructed a quantitative microbial risk assessment model as a Bayesian belief network (BBN), a type of graphical model used to support causal inference. The BBN characterizes the concentration of Ebola virus at potential locations where wastewater and sewer workers could be exposed. The BBN estimates risks in a prototypical wastewater system serving a mid-sized city for three different worker exposure scenarios: (1) inhalation of water droplets over a four-hour period, (2) ingestion assuming a worker falls into wastewater and is immersed for 10 seconds, and (3) ingestion assuming one-minute immersion. Based on the calculated virus concentration at various locations within the sewer system or treatment plant plus the assumed exposure scenario, the BBN calculates an expected Ebola dose (in units of number of viruses inhaled or ingested). Then, using data from previous studies of Ebola virulence, the BBN calculates a probability that the worker will become ill.
To identify guidelines for managing liquid waste from hospitalized Ebola patients during the 2014 Ebola outbreak, we reviewed protocols developed during the epidemic by the WHO, U.S. federal and state agencies, and selected U.S. hospitals and associations. The review documented wide variation in recommended practices. Importantly, while WHO and CDC guidelines endorse flushing patient waste down toilets, sinks, or shower drains without pretreatment, the U.S. Army Institute of Public Health and 10 of 12 state and local protocols require disinfection prior to flushing. Among the protocols recommending disinfection, suggested disinfectant concentrations and contact times vary, from no recommendations on concentrations or contact time to recommending specific disinfectant volumes and time periods from 5 to 15 minutes. One hospital (Bellevue in New York) requires solidifying patient waste and disposing of it as Category A solid waste. These conflicting guidelines contributed to wastewater utilities’ confusion about the adequacy of existing hospital practices to protect their workers.
To provide a consistent approach for patient liquid waste treatment that can be deployed during future outbreaks of Ebola or other highly virulent pathogens, the 33 expert workshop participants developed consensus guidelines based on their review of existing protocols. The consensus protocol is modeled on the U.S. Army Institute of Public Health Ebola Virus Disease Waste Management in the Medical Treatment Facility document. Participants concluded that although evidence to determine the efficacy of existing protocols is insufficient, the U.S. Army protocol was the most conservative and comprehensive among those reviewed, while remaining feasible to implement. Due to concerns about hospital worker and patient exposure to chlorine fumes, along with evidence of the efficacy of peracetic acid in Ebola inactivation, the workshop participants recommended modifying the Army’s protocol to use peracetic acid as the disinfectant and to provide 15 minutes of disinfection time prior to flushing.
Overall, under realistic exposure scenarios, our risk model indicates that the risk of sewer and wastewater workers becoming infected with Ebola virus in the event that an Ebola patient is being treated in a hospital that discharges to a sewer system is extremely low. Only when the worker ingests wastewater at volumes comparable to that ingested during a 10-second or one-minute swim is the risk significantly greater than zero (Table ES-1, top seven rows). The maximum daily risk under a worst-case scenario in which a sewer worker near the hospital outlet is immersed in wastewater for one minute is 1 in 450 (Table ES-1, top row); the monthly risk in this worst case is 1 in 23.
Although estimated risks are very low under most realistic exposure scenarios, construction of this risk model has revealed that workers could be at risk under some scenarios and that fully understanding the potential for such worst-case scenarios to occur requires better information about work practices. Because data on volumes of wastewater that workers may inhale or ingest are unavailable, we had to assume exposure volumes comparable to those occurring during a shower or a short swim. Also unavailable is information about the potential for wastewater ingestion or inhalation when conducting different work tasks, for example when repairing sewer mains, raking bar screens at the treatment plant inlet, or repairing primary settling tanks or others units within the plant. Worker practices for using personal protective equipment (PPE) that may mitigate such exposure also are poorly understood. Our sensitivity analysis shows that the estimated risk is highly sensitive to assumptions about exposure volumes. As a result, gathering information about potential exposure volumes and PPE practices for different work tasks is essential for improving future risk models.
|Disinfection Method||Disinfection Exposure Time (min)||Temperature (⁰C)||Distance from Hospital (mi)||Tertiary Treatment||Hospital Size (ft2; In-Hospital Dilution)||% of WW Coming from Hospital||Exposure Method||Daily Infection Risk||
Monthly Infection Risk
Worst Case Scenario
|Severe||None||-||≤15⁰C||≤1||No||≤500k||>95%||1 min ingest||1 in 450||1 in 23|
(Changes from the worst case scenario highlighted in bold type)
|Severe||None||-||>29⁰C||≤1||No||≤500k||>95%||1 min ingest||1 in 475||1 in 24|
|Severe||None||-||≤15⁰C||≤1||No||>1,000k||>95%||1 min ingest||1 in 715||1 in 36|
|Severe||None||-||≤15⁰C||≤1||No||≤500k||>95%||10s ingest||1 in 14,285||1 in 715|
|Severe||None||-||≤15⁰C||≤1||No||>1,000k||>95%||10s ingest||1 in 25,000||1 in 1,250|
|Severe||None||-||≤15⁰C||2-3||No||≤500k||<5%||1 min ingest||1 in 33,333||1 in 1,667|
|Severe||Bleach||2 min||29⁰C||≤1||No||≤500k||5-50%||10s ingest||1 in 50,000||1 in 2,500|
|Severe||Quats||2 min||≤15⁰C||≤1||No||≤500k||>95%||1 min ingest||<1 in 100,000||1 in 5,000|
|Nonsevere||None||-||≤15⁰C||≤1||No||≤500k||>95%||1 min ingest||0||0|
|Severe||None||-||≤15⁰C||≤1||Yes||≤500k||>95%||1 min ingest||0||0|
|Severe||Peracetic acid||2 min||≤15⁰C||≤1||No||≤500k||>95%||1 min ingest||0||0|
Risks to wastewater and sewer workers in the event that Ebola virus is flushed into the sewer system are below 1 in 100,000 unless the worker ingests wastewater comparable to ingestion during short swims. Nonetheless, to protect workers in case of unintentional ingestion, we recommend that hospitals pretreat Ebola patient liquid waste prior to flushing into the sewer, following the protocol shown in Figure ES-1. Our risk model shows that this disinfection practice can eliminate risks to wastewater and sewer workers under all scenarios considered in this study.
Although risks of Ebola infection among wastewater and sewer workers are low, previous epidemiologic studies have documented that this workforce faces higher risks of other infectious diseases than the general population. Insufficient occupational protections for this workforce may contribute to this elevated risk. For example, OSHA requires employers to implement a range of strategies (including use of PPE) to safeguard workers from blood-borne pathogens, but these standards do not apply to sewage-borne pathogens; there are no PPE requirements specific to the wastewater industry. Better understanding how to protect this workforce requires data, currently unavailable, on the volumes of wastewater exposure for different work tasks. Also required is information about current use of PPE among wastewater and sewer workers and the extent to which such equipment can mitigate infectious pathogen exposure risks in this population. The health of sewer and wastewater workers is vital to the function of our communities. This study has revealed the need for better information on how to safeguard this workforce from infectious diseases.
Ebola BBN in Netica (.zip archive)
Ebola BBN user manual (PDF)
Ebola Flushing protocol (PDF)
Water Institute and Affiliated Researchers
Jackie MacDonald Gibson – Principal Investigator
Mark Sobsey – Co-Principal Investigator
Jamie Bartram – Co-Principal Investigator
Joe Zabinski – Graduate Research Assistant
Kelsey Pieper – Postdoctoral Research Assistant