2007 Progress Report: Community-Randomized Intervention Trial with UV Disinfection for Estimating the Risk of Pediatric Illness from Municipal Groundwater Consumption

EPA Grant Number: R831630
Title: Community-Randomized Intervention Trial with UV Disinfection for Estimating the Risk of Pediatric Illness from Municipal Groundwater Consumption
Investigators: Borchardt, Mark , Belongia, Edward , Kieke, Burney , Loge, Frank
Institution: University of Wisconsin - Madison , Marshfield Clinic Research Foundation , University of California - Davis , Washington State University
Current Institution: Marshfield Clinic Research Foundation , Washington State University
EPA Project Officer: Hiscock, Michael
Project Period: January 1, 2005 through December 31, 2007 (Extended to December 31, 2009)
Project Period Covered by this Report: January 1, 2007 through December 31,2008
Project Amount: $2,289,169
RFA: Microbial Risk in Drinking Water (2003) RFA Text |  Recipients Lists
Research Category: Drinking Water , Water , Health Effects


  1. Estimate the attributable risk for acute gastrointestinal illness (AGI) and febrile illness from drinking municipal water in communities that use non-disinfected groundwater.
  2. Partition the attributable risk for AGI and febrile illness into two components, risk related to contaminated source waters and risk related to water distribution systems.
  3. Determine if there is an association between virus concentration in water (e.g., adenoviruses 40/41, coxsackievirus, echovirus, and noroviruses) and community illness rates.

Progress Summary:

Years 3 and 4 of the project were spent primarily on four major activities: 1) Implementing the crossover component of the UV disinfection intervention; 2) Collecting drinking waters samples and conducting disease surveillance for two 3-month periods; 3) Analyzing water samples for waterborne human viruses; and 4) Commencing data analysis.

In the beginning of 2007, during the months of January and February, the UV disinfection cross-over was implemented. In other words, the 17 UV reactors installed on 17 wellheads in the eight communities subject to the UV intervention in 2006 were removed, transported, and installed on 17 wellheads in the six communities that had been controls in 2006.  The well houses and piping of the 2006 intervention communities were restored to their original condition and configuration. These communities then became controls for the 2007 data collection effort. The UV installations and startups on the wellheads of the six 2007 intervention communities proceeded as planned and without major problems. All UV lights were operational and meeting disinfection specifications in time for the start of the third surveillance period in March 2007. It was a major accomplishment to complete the cross-over implementation as scheduled, and moreover, ensure that all 14 study communities were pleased with the transition and remained in the study. During the two years of intensive data collection, not a single community dropped-out of the study.

The third surveillance period was launched on March 5, 2007, and continued for 12 weeks until May 27, 2007. As in the previous year, the study families every week completed their symptom checklist and mailed them to the study team. Health data was entered into a database as it arrived daily and all study families were tracked in real-time for their compliance in submitting the symptom checklists. Families that were one week late with their submissions received a phone call reminder from a study coordinator. Significant effort was spent ensuring that all checklists were returned in a timely manner.

Water samples for virus analyses were collected three times, approximately monthly, from each of the 14 study communities during the 12-week period. Samples were taken from the drinking water wells, immediately following UV disinfection at the wellhead if the community was subject to the UV intervention, and at six to eight households per community to represent the distribution system water quality. All samples were processed and archived for later analysis.

During 2007, as in 2006, the water superintendents of the 14 communities were asked periodically to complete a written survey in which they reported operation and maintenance procedures and unanticipated events that occurred in the drinking water distribution system of their community.  Daily pumping volumes from each well were also reported. All data were entered in the study database.

The fourth and last 12-week surveillance and water sampling period began September 3, 2007, and ended November 25, 2007. Every study subject submitted a saliva specimen at the beginning of the period and every four weeks thereafter for a total of four specimens per study subject. Combined with the saliva specimens collected during Period 2, we have archived 8,072 specimens, 5,075 from children and 2,997 from adults. Of these, 7,551 (93.5%) were frozen within 3 days of collection, suggesting ideal preservation of antibodies. Of the 1,372 subjects that submitted saliva, 1,087 (79.2%) submitted four or more consecutive specimens, a time-series very conducive for detecting sero-conversion. Depending on funding availability, the saliva specimens will be analyzed for enterovirus seroconversion, another health outcome measure that could be associated with drinking water exposures. 

Removal of the UV disinfection reactors from the six intervention communities began in December 2007 and was completed by February 2008. All communities reported they were satisfied with the restoration of their wells to their original condition. One study community, Prairie du Sac, purchased two of the UV reactors and after review and approval from the Wisconsin Department of Natural Resources these reactors were permanently installed on the community’s two wells. The residents are now being supplied with UV disinfected drinking water without any chlorination. The remaining 15 UV reactors are cleaned and crated and in storage, ready for sale. Funds from the sale to Prairie du Sac and future sales will be used to support the research activities related to completing the project.

As originally proposed, over the 2 years of 2006 and 2007 the study team successfully completed four 12-week AGI surveillance and water sampling periods. The number of water samples collected for virus analyses is unprecedented, 2,054, from the wells and distribution systems of the 14 communities. The study subjects submitted a total of 64,264 weekly symptom checklists and, as mentioned earlier, 8,072 saliva specimens. Defining enrollment as having submitted at least one weekly symptom checklist, enrollment at the beginning of the study was 621 households with 580 adult and 1079 child study subjects. The number of households per community ranged from 28-70. Twenty months later, when surveillance was completed, the number of households still enrolled was 440 with 413 adults and 765 children, a 29 percent drop-out in households over the data collection period. Characteristics of enrolled households and study subjects are reported in Tables 1 and 2, respectively.

Table 1. Characteristics of households Table 2. Characteristics of study  subjects enrolled in the Wisconsin WAHTER Study enrolled in the Wisconsin WAHTER Study
Characteristic Number %
Household size (no. of persons)
2 17 (3)
3 159 (26)
4 246 (40)
5 136 (22)
>6 63 (10)
Residence type
Single family home 572 (92)
Apartment or condo 43 (7)
Other 6 (1)
Faucet or plumbing filtering device
Yes 73 (12)
No 547 (88)
Don’t know 1 (<1)
Primary drinking water source
Municipal 1546 (93)
Bottled water 58 (3)
Other 1 (<1)
Missing 54 (3)
Characteristic Number %
Age (years)
< 2 147 (9)
3-5 277 (17)
6-12 575 (35)
13-30 193 (12)
31-50 440 (27)
>50 27 (2)
Gender Adults
Male 107 (18)
Female 473 (82)
Gender Children
Male 524 (49)
Female 555 (51)
White 1550 (93)
Non-White 96 (5)
Missing 13 (1)

In Year 4 of the project, 2008, the primary activities were analyzing the water samples for viruses and data analysis. The virus analyses include conducting real-time qRT-PCR/qPCR for six virus groups: enteroviruses, adenoviruses, rotavirus, hepatitis A virus, and norovirus genogroups I and II. In addition, all samples positive by PCR for enteroviruses and adenoviruses are tested for infectivity by cell culture, and the enterovirus or adenovirus serotype is determined by nucleic acid sequencing. To date, all qRT-PCR and qPCR assays have been completed. Infectivity assays for enteroviruses have been completed; the adenovirus assays will be completed within the next month. Serotype identification of enteroviruses and adenoviruses is approximately 80% complete and should be finished in a month.

As for data analysis, the degree of progress depends on the study objective. Objective 3, associating tap water virus concentrations with community AGI rates, is the furthest along and the manuscript is in preparation. Data analyses for the other objectives are progressing. It is anticipated that all analyses will be completed by the December 2008, when the WAHTER Study team is scheduled to present the study findings at the EPA Symposium on Groundwater-Borne Infectious Disease, Etiologic Agents and Indicators.

As part of the work necessary for accomplishing the primary study objectives, the study team developed two new methodological approaches that are broadly applicable to water-related microbiology and epidemiology. Concentrating viruses from drinking water requires specialized filters that are very expensive. Relying on previous reports in the literature, we developed and validated an inexpensive filter constructed with glass wool for concentrating viruses from the thousands of drinking water samples collected in this study. Our validation data were published in Applied and Environmental Microbiology earlier this year. As another cost and time-saving measure we developed new mathematical approaches for estimating the quantity of infectious viruses in a water sample inoculated into a cell culture. The approaches rely on characterizing the growth kinetics of the environmental virus isolate, which takes less time and fewer lab supplies than quantification by most-probable-number. Unfortunately, development took longer than anticipated and to complete all analyses within the study timeframe the method was not applied to samples collected for this study. The method is described in a manuscript entitled, “New Mathematical Approaches to Quantify Human Infectious Viruses from Environmental Media Using Integrated Cell Culture-PCR” and is currently in review.

Future Activities:

The study team has requested from EPA a no-cost extension for this project. Activities during this time will focus on completing data analyses and writing several manuscripts for publication. Specifically, with regard to data analysis, we will evaluate the intervention effect of UV disinfection in reducing pediatric illnesses, estimate the fraction of illness risk attributed to contaminated groundwater versus drinking water distribution systems, and identify operation and maintenance events in distribution systems that are associated with virus contamination. With regard to manuscript writing, we will prepare four more publications: 1) The association between household tap virus concentrations and community rates of acute gastrointestinal illness (AGI); 2) Attributable risk of waterborne AGI from contaminated groundwater (i.e., the UV intervention effect); 3) Attributable risk of AGI from virus intrusions into drinking water distribution systems; and 4) The association between operation and maintenance procedures in distribution systems and virus contamination. The study team expects to be highly productive during this time period.

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 26 publications 1 publications in selected types All 1 journal articles
Type Citation Project Document Sources
Journal Article Lambertini E, Spencer SK, Bertz PD, Loge FJ, Kieke BA, Borchardt MA. Concentration of enteroviruses, adenoviruses, and noroviruses from drinking water by use of glass wool filters. Applied and Environmental Microbiology 2008;74(10):2990-2996. R831630 (2007)
R831630 (2008)
R828041 (Final)
  • Abstract from PubMed
  • Full-text: AEM
  • Other: AEM PDF
  • Supplemental Keywords:

    Epidemiology, human health, pathogens,, RFA, Scientific Discipline, INTERNATIONAL COOPERATION, Water, Environmental Chemistry, Health Risk Assessment, Drinking Water, Environmental Engineering, microbial contamination, microbial risk assessment, monitoring, real time analysis, aquatic organisms, other - risk assessment, early warning, UV disinfection, children, municipal groundwater, water quality, drinking water contaminants, drinking water system

    Relevant Websites:

    Marshfield Clinic Research Foundation Exit
    Principal Investigator Profile Exit

    Progress and Final Reports:

    Original Abstract
  • 2005 Progress Report
  • 2006 Progress Report
  • 2008 Progress Report
  • Final