Grantee Research Project Results
2005 Progress Report: A Prospective Epidemiological Study of Gastrointestinal Health Effects Associated with Consumption of Conventionally Treated Groundwater
EPA Grant Number: R830376Title: A Prospective Epidemiological Study of Gastrointestinal Health Effects Associated with Consumption of Conventionally Treated Groundwater
Investigators: Moe, Christine L. , Rose, Joan B. , Moll, Deborah , Payment, Pierre
Current Investigators: Moe, Christine L. , Moll, Deborah , Nilsson, Kenneth , Hooper, Stuart
Institution: Emory University , INRS - Institut Armand-Frappier , Centers for Disease Control and Prevention , Michigan State University
Current Institution: Emory University , Centers for Disease Control and Prevention , University of South Florida
EPA Project Officer: Page, Angela
Project Period: October 1, 2002 through September 30, 2005 (Extended to September 30, 2008)
Project Period Covered by this Report: October 1, 2004 through September 30,2005
Project Amount: $1,820,900
RFA: Microbial Risk in Drinking Water (2001) RFA Text | Recipients Lists
Research Category: Water , Drinking Water , Human Health
Objective:
The objectives of this research project are to: (1) estimate the risks of endemic gastrointestinal illness (GI) associated with the consumption of conventionally treated groundwater in the United States; and (2) determine the relative contributions of source water quality, treatment efficacy, and distribution system vulnerability to endemic waterborne disease. This study involves a 12-month, double-blinded, randomized drinking water intervention trial with 900 households that will measure rates of GI drinking water groups that receive different levels of treatment.
Approach:
The proposed study is a 12-month, double-blinded, randomized intervention trial of 900 households in a large metropolitan area in the southeastern US with a community GW system that uses conventional treatment methods, meets current water quality standards and has a well-characterized distribution system with areas of vulnerability. Study households will be randomly divided into three groups of 300 households: Group 1 = Households that drink bottled water that has been treated with reverse osmosis and other advanced treatment methods, Group 2 = Households that drink bottled municipal water collected at the water treatment plant, and Group 3 = Households that drink municipal tap water. Groups 1 and 2 will be blinded to their group assignment. Half of the households within each group will be recruited from vulnerable areas in the distribution system in order to examine the health risks from distribution system contamination. Study participants will report GI symptoms and selected risk factors in a weekly health diary and biweekly telephone interview. Samples will be routinely collected from raw source water, treated water, distribution system water and bottled water and analyzed for microbial indicators. Data analyses will compare GI rates and water quality between the three study groups and between study households in different parts of the distribution system.Progress Summary:
Status of Project
In this reporting period, the project focused on subject recruitment and enrollment. Recruitment of qualified households has been extremely difficult, as is discussed in detail below. We currently recruited 286 households from phone contacts made with about 30,000 households in the study area. This represents 26% of our target recruitment of 1,100 households. Study staff are currently enrolling recruited households. The purpose of the enrollment visit is to explain the study to all household members, confirm eligibility to participate, collect baseline specimens, provide study materials (health diaries, etc.), and obtain informed consent.
Follow-up contact by study personnel and field nurses to schedule a household visit found that 26% of the 286 recruited households were not correctly classified as qualified during the telephone interview or no longer qualify due to a change in household circumstances (such as recent purchase of a home water filter), and 28% are no longer interested in participating in the study. About 39% are still interested and will be visited for final enrollment.
2.1 Recruitment
A. Phase 1 Recruitment by Letter and Telephone Follow-up: March –April 2004
A list of water customers, addresses, and phone numbers was obtained from Hillsborough County Water Department. A mailing including a full-color brochure explaining the study was prepared for 14,000 households in the Hillsborough County database. A local printing agency was contracted to create the brochure design, print, address, and stuff the 14,000 envelopes. Study personnel finished the mailing, following US Postal Service guidelines for bulk mailing.
Prior to mailing, the list was checked for accuracy and to ensure addressing conformed to US Postal Service Standards by an independent company, Accurate Leads. In doing so, it was determined that many of the phone contact numbers provided by Hillsborough County did not match those provided by Accurate Leads. It was later determined that the numbers provided by Hillsborough County were often not accurate, as many were contact numbers given at time of start of service (cell phone, work, etc.) and were not maintained over time. Accurate Leads was also able to provide us with names, addresses, and phone numbers for households living in apartment buildings and similar developments, which were not included in the Hillsborough County database.
Eligible households for this study are those whose members drink mostly tap water, without using any filters or softeners. Households had to be located within the study area, be receiving municipal tap water, and not have any major health issues, among other criteria. A detailed script was developed to determine the eligibility status of households contacted. An initial requirement that households have at least one child between the ages of 2 and 16 was dropped due to the low number of households that met our eligibility criteria.
The project contracted in February with an Atlanta-based telemarketing firm (RDI Telemarketing, Inc.) to perform telephone recruitment. In consultation with Emory University personnel, RDI developed a script for the calling campaign. After 7,000 letters were sent out (Phase 1) in March, 2004, RDI began calling those who received these letters on April 1, 2004. After 2 weeks of calling, the campaign was suspended due to low numbers of eligible households. Based on extrapolating the rate of return, it was clear the study would not reach target enrollment based on the 14,000 household list.
B. Community Outreach Campaign
A community outreach campaign was developed to raise awareness about the study within the Hillsborough County community. A website was developed and a press release issued. Working in conjunction with media relations at University of South Florida (local to the study area), contacts were made at various print, radio, and television media. The following organizations were contacted directly for their collaboration in recruitment:
- Florida Department of Health
- College of Public Health at the University of South Florida
- Hillsborough County Health Department
- Hillsborough County Office of Neighborhood Relations
- Hillsborough County Public Library System with 11 libraries and affiliated libraries
- University of South Florida Health Sciences Center
- University of South Florida Department of Publications and University Relations
- University of South Florida Health Sciences Public Affairs
- Redlands Christian Migrant Association
- Hillsborough County Home Owners Associations
- All day care centers in the study area
- Churches in the study area
In addition, study posters, brochures, and flyers were posted at seven summer camps and YMCAs, 20 stores and supermarkets, 44 day care centers, six gyms, ten hospitals and clinics, ten laundromat stores, one cyber cafe, two main health sciences fairs (with booth at one), 46 apartment complexes, the main University of South Florida (USF) library, the Marshal Center at USF, the Moffit Cancer Center, and the USF Gym. In addition, 94 churches were contacted by phone and 23 were visited by our field staff in order to leave study brochures.
C. Media Exposure
With permission from Hillsborough County Water Department officials, the study pursued local media exposure. This effort resulted in articles in eight local newspapers, with three separate articles printed in The Tampa Tribune. In addition, three magazines have featured the study. A public service announcement describing the study was featured on several local radio stations and University of South Florida Radio (WUSF) interviewed Dr. Izurieta and Dr. Stark. The study was featured on two stations during the evening local news program, Bay News Channel 9 and ABC News Channel 28. In addition, Channel 22 (HTV) interviewed Dr. Izurieta for a story and a documentary about the study was produced and presented by Channel 8.
Hillsborough County Office of Neighborhood Relations Newsletter: May/June 2004
St. Petersburg Times: June 4, 2004
HTV Channel 22: June 18, 2004
Bay News 9 TV: June 29, 2004
The Tampa Tribune: June 29, 2004
Water Tech Online: June 29, 2004
WUSF 89.7 Radio: July 4-July 10, 2004
USF News - Focus on Health Sciences: July 15, 2004
Hillsborough County Communicator: July 2004 (insert in all water bills)
ABC News Channel 28: July 27, 2004
Osprey Observer: July 2004
The Tampa Tribune, Brandon Section: August 7, 2004
Brandon-Valrico Journal: August 2004
The Tampa Tribune: August 18, 2004
Senior Connection: August 2004
Osprey Observer: August 2004
Water Europe (Journal of the European Bottled Water Association): Autumn 2004
The Tampa Tribune: September 12, 2004
D. Phase 2 Telephone Recruitment: June 15 – July 27
An additional list of study area residents was purchased from Accurate Leads, increasing our database from 14,000 to 60,000 households. On June 15, 2004, telephone recruitment resumed. After 4 weeks of calling, the rate of return resulted in a projected 820 eligible households. RDI was asked to increase the call volume, and on July 12 the call volume was increased by a factor of eight. However, a decline in eligibility rate from 5% to 0.5% was associated with the increase in call volume. This resulted in a much lower return of actual qualified households. The reasons for this drop in the eligibility rate during the ramp-up period are being investigated by RDI. Figure 1 below shows the daily qualification (eligibility) rate (black) and the running average qualification (eligibility) rate (blue) over the recruitment period. After this ramp-up period, the call volume was reduced again to allow us to analyze the results while still attempting to recruit study households at a slower rate.
Figure 1. Proportion of Eligible Households Recruited by the RDI Telephone Campaign: June – September 2004
2.2 Summary of Progress
- Supplies for 1,100 household kits, 4,500 subject kits, and 2,200 stool sample collection kits were ordered. The kits included consent forms, household enrollment forms, individual enrollment forms, bi-weekly health diaries for each household member, water bottles for each household member, blood collection tubes and phlebotomy supplies, saliva collection tubes and lemon drops, stool collection cups, instructions for collecting saliva and stool samples, instructions for filling out the health diaries, VOC water sample bottle, water exposure questionnaire, refrigerator magnet with study 1-800 telephone number. The kits were assembled at Emory by a team of student assistants and shipped to a storage facility in the study area.
- A study logo was developed. This logo was incorporated into the full-color brochure, magnets with the study hotline number, and water bottles. The magnets and water bottles were designed and ordered.
- Study website was established.
- All consent forms, brochures, questionnaires, health diaries, and other study instructions were finalized. All documentation was translated into Spanish by the project team. (Four Spanish/English bilingual speakers work on the project.) These documents include the Household Questionnaire, Subject Questionnaire, Water Consumption Questionnaire, Bi-weekly Health Diary, Consent Forms (7), Letter to Recruited Households, the Brochure, and the Telephone Survey.
- Made 44,693 telephone contacts (including bad numbers) to residents of the study area to determine interest and eligibility to participate in the study. Initially, a total of 286 households were recruited (more information below).
- Developed a media exposure campaign (described above).
- Project has contracted with the Florida Department of Health Laboratory for analyses of the water samples collected during the study. This work will be supported by a contract from AWWARF directly to the Florida Department of Health – Bureau of Laboratories.
- Field office space was secured: one office with two desks, two computers, one scanner, one fax machine, and internet connection at the University of South Florida - College of Public Health.
- Field laboratory space secured: space includes access to -80°C freezer, refrigerator, and standard freezer, all at University of South Florida College of Public Health.
- Storage space secured in area central to the study area.
- A database containing data entry forms, tables, and visual basic programming have been designed for quality assurance of the questionnaires.
- Quality assurance programming using SAS, including creation of SAS double entry discrepancies code and a check errors procedure for all questionnaires.
- Developed procedures for data collection and transfer between field office and the Florida Department of Health Laboratory
2.3 Preliminary Results and Evaluations
A. Recruitment of Study Households
Homes called during the telephone recruitment campaign generally fall into four main categories: 1) bad telephone number, 2) not interested, 3) not eligible based on study criteria, and 4) eligible and interested. Table 1 below summarizes these outcomes:
Table 1. Outcome of Telephone Recruitment Calls: April – September 2004
Outcome | Number |
Percent of Total |
Percent of Total Contacts |
Bad Number |
10,195 |
22.8% |
N/A |
Not Interested |
19,579 |
43.8% |
56.8% |
Not Eligible |
14,698 |
32.9% |
42.6% |
Interested and Eligible |
221 |
0.5% |
0.6% |
Total Attempted Calls |
44,693 |
||
Total Contacts* |
34,498 |
*Total Contacts includes all households where we were able to speak to someone. Bad telephone numbers are not included as contacts. Telephone numbers that were busy, had no answer or had an answering machine were recalled.
An additional 65 households were qualified for the study based on inbound calls to the study 800 number. Of all households contacted, about 43% expressed interest in the study, at least to the extent that one or more eligibility questions were answered to determine that the household was not eligible. Of all interested households, 98.5% did not meet the eligibility criteria to participate in the study. Table 2 summarizes reasons why households were not eligible to participate in the study based on outbound telephone recruitment.
Table 2. Reasons why contacted households were not eligible to participate in the FRESHWATER Study: April – September 2004
Reason | Number |
Percent |
Drink Filtered Water |
8,446 |
56.6% |
Drink Home Softened Water |
1,079 |
7.2% |
Drink Bottled Water |
4,375 |
29.3% |
Drink Boiled Water |
15 |
0.1% |
Other* |
798 |
5.2% |
Eligible |
221 |
1.5% |
Total |
14,919 |
* Other: includes households with a significant health problem, using a private well, not living in the study area, not planning to remain in the study area for the next year, etc.
More than half of disqualifications were due to use of a filter for household drinking water, which included whole house filtration units, under-the-sink filtration units, faucet filters, water jug filters ("Brita" type and similar), and refrigerator filters.
The telephone recruitment effort yielded information on drinking water habits of the residents of the study area contacted. Table 3 summarizes the drinking water habits of households who were contacted during the telephone campaign.
Table 3. Reported drinking water habits of Hillsborough County households: April – September 2004
Drinking Water Habit | Number |
Percent |
Drink Filtered Water |
8,446 |
59.3% |
Drink Home Softened Water |
1,079 |
7.6% |
Drink Bottled Water |
4,375 |
30.7% |
Drink Boiled Water |
15 |
0.1% |
Drink Tap Water* |
319 |
2.2% |
Total |
14,234 |
*The number of households drinking tap water is greater than the final number of households that were eligible to participate in the study because some households that qualified for the study based on their drinking water habits were disqualified by other health-related eligibility criteria.
Since the survey was not designed to determine drinking water habits only, these numbers are only approximate. For example, households who did not live in the study area were disqualified before any drinking water habit questions were asked. However, the numbers of other disqualifications were relatively small.
As the recruitment campaign progressed, the telephone screening survey was modified to retrieve more information households drinking filtered water. These results are summarized in Table 4.
Table 4. Reported use of household water filters: Hillsborough County, 2004
Type of Household Water Filter | Number |
Percent |
Faucet Filter |
104 |
18.4% |
Under Sink Filter |
53 |
9.4% |
Refrigerator Filter |
192 |
33.9% |
Water Pitcher Filter |
29 |
5.1% |
Whole Household Filter |
102 |
18.0% |
Other |
67 |
12.2% |
Total |
547 |
These data indicate that most households with a water filter used a refrigerator dispenser. Faucet filters and whole house filters were also very popular.
B. Vulnerability Analyses of the Hillsborough County Water Distribution System
The vulnerability analyses described in the first annual report was completed. The analyses included the following data sources:
Parameters that Affect Water Quality
Outcome Measures
- Heterotrophic plate count bacteria (HPC)
- Turbidity
- Positive total coliform samples
Impact of Operations and Infrastructure
- Residence time
- Scheduled flushing events
- Repump facility water
Parameters that Affect the Potential for Intrusion and Cross Contamination
- Pressure during normal demand operation
- Critical pressure during a simulated power outage event
- Proximity of water lines to septic tanks
- Proximity of water lines to sewer mains
- Proximity to reclaimed water use
- Main breaks and repairs
The analyses resulted in an estimated vulnerability score for every distribution system node (juncture of pipes). Some of these data, such as distribution of pressure and residence time, were derived from running the distribution system model using EPANET. Other data, such as turbidity and total coliform positive samples, were mapped by sample location and EPANET was used to simulate a conservative tracer addition simultaneously at each sampling location. The model was supplied by the utility and is one that is used by the utility on a regular basis.
Each parameter analyzed resulted in a "layer" of mapped values across the system. The values were normalized and then summed, with higher values indicating greater vulnerability. Finally, the nodes with the lowest and highest 20th percentile in summed value were designated the less and more vulnerable areas, respectively. In Figure 2 below, blue nodes represent less vulnerable areas while red nodes represent more vulnerable areas of the system.
Figure 2. Estimated areas of less vulnerability and more vulnerability in the Hillsborough County Water Distribution System.
3. Problems and Difficulties Encountered
3.1 Denied requests to recruit via Hillsborough County School System
The original study recruitment plan was based on identifying families with children by obtaining addresses of school children in Hillsborough County. We would then follow up with telephone screening calls to determine the interest and eligibility of these families for our study. The County has 30 elementary schools serving approximately 24,500 students and 9 middle schools serving approximately 11,400 students. Our study coordinators spoke to a number of school principals and school nurses who were interested in the study, but we were informed that we needed to get permission from the Hillsborough County School Board before addresses of students could be released. In late September 2003, our first request to the School Board was denied on the grounds that it violated the US Department of Education Family Education Rights and Privacy Act (FERPA). We then asked the School Board if we could send study information brochures home with school children in Hillsborough County because this is done for other research studies in the Atlanta area. This request was denied in May 2004.
3.2 Low proportion of households drink tap water
The main difficulty encountered during the reporting period was the low proportion of households that met the eligibility criteria for the study. Only 0.6% of households that were contacted were eligible to participate in the study. Consequently, the recruitment period has lasted longer than anticipated and has not yielded the necessary number of study households. The main reason for the low qualification rate is that most residents of the study area do not drink unfiltered tap water. Many households cite odor and taste as the primary reason why they use a filter or drink bottled water. This groundwater system has a sulfur taste, and many people in the Tampa area have moved there from other parts of the country and are not used to the taste of the water. A 2003 survey of 400 City of Tampa Water Department customers indicated that 28% of the customers had moved to the Tampa area in the past 5 years and 53% had lived in the area for <10 years. This survey also indicated that about 35% of these customers regularly drank bottled water, and 82% of bottled water drinkers reported that they drank bottled water because of taste. (Note: The City of Tampa Water Department uses a surface water supply – NOT the groundwater supply used by Hillsborough County Water Department.)
Another potential problem that may have reduced the number of tap water drinkers in Hillsborough County is that the utility had at least two Total Coliform Rule violations during and just prior to the recruitment period.
3.3 Consumers wary of phone calls asking about drinking water
We have also learned that there has been aggressive telemarketing of bottled water and home water treatment devices in the Tampa area. This may make residents wary of telephone calls asking about drinking water habits and recruiting for a research project about drinking water. The Office of the Florida Attorney General issued a warning to alert consumers about fraudulent offers for “free water testing” in order to sell bottled water or water treatment devices (see appendix). In an effort to distinguish our telephone recruitment from other telemarketers, our callers used a standard script that emphasized that this was a research study (see appendix) and “Emory University” appeared on caller ID screens.
3.4 Limited interest in participating in research study with no incentives
We found that some households were simply not interested in participating in a research study where there was no perceived benefit. During the telephone screening interview, the household was told that they have a 1 in 3 chance of continuing to drink tap water and a 2 in 3 chance of receiving bottled water. The only incentives that we were able to offer at this time were sports water bottles with the study logo, the refrigerator magnet with the study logo and a couple product samples donated by Proctor and Gamble. We plan to solicit local businesses for donations of goods and services that we can offer our study families in a monthly raffle. However, because we have been putting all our efforts into recruiting study families and launching the study, we have not had time to solicit donations.
3.4 Severe weather in study area hampered recruitment
Telephone recruitment was suspended for three weeks during the summer/fall 2004 when four hurricanes/tropical storms in a short period affected the study area: Charley August 13, Frances September 5, Ivan September 17 and Jeanne September 25 (see appendix). Some parts of the study area had no power for an extended period of time. Schools were temporarily closed, and there was flooding in some areas.
Expected Results:
This is the first study to measure the risk of GI associated with the consumption of conventionally treated GW and to distinguish between the risk from source water and treatment vs. the risk from the distribution system. The results of this study will provide valuable information on the magnitude of endemic GI associated with drinking water in the US.Future Activities:
Given the results of the data from the telephone recruitment campaign, it is clear that the study will need to be modified before future recruitment efforts are attempted. Please see the interim progress report that describes study activities from October 2004 through May 2005 and summarizes the changes to the study design, the results of the new recruitment campaign and proposed alternate study designs.
References:
Friedman M, Radder L, Harrison S, Howie D, Britton M, Boyd G, Wang H, Gullick R, LeChevallier M, Wood D, Funk J. Verification and control of pressure transients and intrusion in distribution systems. American Water Works Association Research Foundation, 2004.
Journal Articles:
No journal articles submitted with this report: View all 12 publications for this projectSupplemental Keywords:
water, drinking water, distribution system, exposure, risk, health effects, human health, pathogens, epidemiology, modeling,, RFA, Health, Scientific Discipline, Water, Environmental Chemistry, Health Risk Assessment, Epidemiology, Risk Assessments, Biochemistry, Drinking Water, groundwater disinfection, health effects, microbial contamination, bacteria, human health effects, waterborne disease, other - risk assessment, exposure, microbial effects, treatment, human exposure, microbial risk, water disinfection, groundwater contamination, water quality, dietary ingestion exposures, drinking water contaminants, drinking water treatment, human health, gastrointestinal health, groundwater, gastrointestinal health effects, exposure assessmentProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.