An Overview of the Be Well Home Health Navigator Program to Reduce Contaminants in Well Water: Design and Methods

Abstract

Background:

The Be Well Home Health Navigator Program is a prospective, randomized controlled trial (RCT) implemented to compare a community health navigator program to usual care program to reduce contaminants in drinking water.

Design and Setting:

This 4-year two-armed RCT will involve well owners in Oregon that have private drinking water wells that contain arsenic, nitrate, or lead above maximum contaminant levels.

Intervention:

The intervention leverages the trusted relationship between Cooperative Extension Service (CES) Community Educators and rural well owners to educate, assist and motivate to make decisions and set actionable steps to mitigate water contamination. In this study, CES will serve as home health navigators to deliver: 1) individualized feedback, 2) positive reinforcement, 3) teach-back moments, 4) decision-making skills, 5) navigation to resources, 6) self-management, and 7) repeated contact for shaping and maintenance of behaviors. Usual care includes information only with no access to individual meetings with CES.

Measurable outcomes:

Pre-specified primary outcomes include 1) adoption of treatment to reduce exposure to arsenic, nitrate, or lead in water which may include switching to bottled water and 2) engagement with well stewardship behaviors assessed at baseline, and post-6 and 12 months follow-up. Water quality will be measured at baseline and 12-month through household water tests. Secondary outcomes include increased health literacy scores and risk perception assessed at baseline and 6-month surveys.

Implications:

The results will demonstrate the efficacy of a domestic well water safety program to disseminate to other CES organizations.

1. The Rationale for the Be Well Home Health Navigator Program

Approximately 44 million Americans rely on private wells for their drinking water, and about 25% of these wells contain contaminants at levels that exceed the maximum contaminant level goals of the Safe Drinking Water Act (SDWA)(1, 2). Arsenic and nitrate are the most frequently detected contaminants in well water (3, 4). Arsenic enters groundwater by the natural decomposition of volcanic materials and as a byproduct of mining and past pesticide use. Nitrate enters the groundwater as a result of runoff or leakage from fertilized soil, waste water, or septic systems. Lead can leach out of plumbing (5-7). The health risks associated with chronic ingestion of arsenic, nitrate, and lead are well documented and include increased risk of cancer, cardiovascular disease, neurocognitive disorders, and adverse reproductive health outcomes (8-20). Private wells are excluded from the SDWA, save for crisis situations that endanger public health. Consequently, well owners shoulder the responsibility of managing the safety of their water. Some organizations offer voluntary programs that distribute risk information, offer hotline support, and occasionally subsidize the cost of water testing (21, 22). While these programs can increase awareness of local hazards and improve testing rates, they may not promote behavior changes that improve well water quality and are often only used by people with higher socioeconomic status which can exacerbate health inequities (23). Prompting behavior change among well owners can be challenging because these contaminants lack sensory cues that are often used to shape behaviors (24, 25). This observation is reinforced by surveys of well owners where the primary reason for not testing or treating well water was that it smelled, looked, or tasted fine (26-30). The long latency period between exposure and adverse health effects also presents challenges for prompting behavior change (24, 31). To address these challenges, a few states have implemented policies that require testing private wells during real estate transactions (32). However, most states rely on voluntary programs that may not overcome these barriers.

Several programs have focused on improving well water testing rates through community interventions. In Renaud et al (33), a mass media campaign and community intervention resulted in a testing rate four times what was observed prior to the intervention. In New Hampshire, Paul et al (34) offered a well water testing service for residents to make testing accessible and reduce inconvenience. This program collected triple the number of water samples tested at the same lab in the previous six years. Fewer programs have delivered interventions to assist with implementing treatment for water contaminants. The Strong Heart Water Study (SHWS) was an intervention to reduce arsenic exposure among private well users in the Great Plains Indian Nations and sought to overcome barriers by offering free water testing and point of use (POU) filters on kitchen faucets (35). They provided a combination of free filters, phone calls, in-home visits and Facebook messaging to encourage the use of POU filters. The program significantly increased perceived vulnerability to arsenic exposure, self-efficacy, and norms (36). At long-term follow-up (average 1.7 years), the majority of participants (93%) had filter faucet concentrations of arsenic below the maximum concentration of 10μg/L (29). A key component to the Strong Heart Water Study was the provision of a free water filter. Our team wanted to develop and test a program that did not require the provision of a free filter in order to increase the economic feasibility of disseminating and scaling up the program.

The Be Well study is a collaborative effort between Oregon State University College of Health and Cooperative Extension Service. The objective of the study is to conduct a RCT that assesses the efficacy of an intervention delivered by trained CES Community Educators to lower the level of arsenic, nitrate, or lead detected in well water. We will randomly assign individual households to one of two study arms: 1) usual care that provides educational materials and 2) the same materials from usual care plus access to a community health navigator. We anticipate more implementation of appropriate treatment and maintenance of well water among individuals assigned to a community health navigator compared to those who receive educational material alone.

2. Methods

2.1. Formative research that informed the Be Well Home Health Navigator Program

No programs currently exist in Oregon that provide well owners with well water treatment options, assist with decision-making, or counsel to overcome barriers to treatment. We collected data through a regional survey of well owners (n=53) about their perception of treatment options and their relative costs. Over 30% of the sample experienced contaminated well water and 18% had water change color or taste, yet 75% did nothing to test or treat their water. Well owners consistently overestimated the cost of treatment option with most quoting figures that would be appropriate for a complete system replacement, rather than making simple system repairs or troubleshooting maintenance techniques.

We also conducted formative semi-structured interviews with 27 community members in Oregon that use private wells. Of these 27 members, 23 interviews were conducted in English and 4 were conducted in Spanish. Interviews were conducted either by faculty investigator who held a PhD, CES Community Educators, or bilingual, bicultural staff with a Master of Public Health. In addition, study team members held conversations at community events that drew rural residents. We learned that they wanted low-cost water testing, improved access and content of educational materials, and connections to a well-water educator. Well owners commented on the lack of resources for learning about water treatment options and were surprised at the variety and affordability of different water treatment options They also expressed a strong distrust of governmental programs and did not want their well water data to be given to governmental agencies. People overwhelmingly trusted the CES to deliver accurate and relevant information.

2.2. Rationale for Navigators

A community health worker (CHW) is a frontline worker who is a trusted member of and/or has a close understanding of the community served to motivate and encourage people (37). A community health navigator (CHN) typically provides coaching and social support specifically getting people to establish care and utilize community resources (37). Both CHWs and CHNs are capable of reaching local populations, providing culturally appropriate health information, and achieving behavior change outcomes among individuals with low literacy (38-41). These were determined to be important components of reaching a rural target audience. Most of Oregon is sparsely populated, and the majority of the counties (26 out of 36) are considered rural (42). We chose CES Community Educators as our “navigators” because of their programmatic experience in groundwater protection and natural resources, their physical locations in rural and frontier counties, and their trust within the community. Established by the Smith Lever Act in 1914, CES provided community-based non-formal education and is embedded within the land-grant universities with staff located in almost every county in the United States (43, 44). Nationally recognized CES programs are SNAP-Ed, Master Gardeners and 4-H Youth Development. CES Community Educators already engage in delivering health education, land management stewardship, and interpretation of test results (45). Materials from two CES programs, Living on the Land and Home-a-Syst, were incorporated into this intervention (46,47).

The Behavioral Ecological Model (BEM) guided our conceptual intervention model and measurement plan (48, 49). The BEM assumes an interaction among physical and social contingencies to explain behavior with a specific emphasis on contingencies of reinforcement across individuals, levels of society, culture, and policy. We selected the BEM because this theory best explains behavior through the cumulative and synergistic effects of 1) policies and government programs, 2) community-level physical environment, 3) community-level culture, 4) community-level environmental resources, and 5) individual characteristics both modifiable and non-modifiable. BEM suggests that incorporation of feedback systems will increase the likelihood that change is achieved. Our intervention would focus on feedback on water test results and changes that can be made to water quality from using appropriate treatment, shaping of social contingencies of well water stewardship practices, and reinforcing treatment and maintenance using health education and motivational interviewing strategies.

2.3. Community Advisory Board.

We convene an 8-member community advisory board (CAB) providing feedback on materials and recruitment process and help to share findings back to the community. This board is comprised of representatives from Oregon Health Authority, Department of Environmental Quality, CES, local health departments, water testing companies, realtors, soil and water conservation districts, and a community member who gets their water from a well. All members except the state representative work or reside in the target area. It meets annually and is given quarterly updates.

2.4. Study Design and Sample Selection

Our intervention is a two-armed RCT, efficacy trial in a community setting (50, 51) A general overview is presented in Figure 1. Study activities were approved by the Oregon State University Institutional Review Board (protocol #2020-0716) The intervention is registered at ClinicalTrials.gov (NCT# 05395663) and we will regularly convene our Data Safety Monitoring Board (DSMB) to review and evaluate accumulated study data for participant safety and to provide recommendations to study investigators.

Figure 1:

Overview of the intervention study design used to evaluate the Be Well Home Health Navigator Program.

Our recruitment strategy is to identify well owners residing in Oregon that plan to live at their home for at least one year and use their well for their household drinking water. We will target recruitment in areas where there was prior knowledge that arsenic or nitrate were elevated in the groundwater and initial recruitment will target Lane, Jackson, and Josephine counties (52). First, we use geographic information systems (GIS) to calculate the average rates of arsenic and nitrate in groundwater aquifers based on data from real estate transactions by Hydrologic Unit Code 12 (HUC12) (32). Then, we identify households with a registered well using publicly available data from the Oregon Water Resource Department (53). We will augment this approach using GIS to identify households outside of known municipal water services areas because there is a significant portion of unregistered wells in these counties.

We prioritize recruitment in areas in census blocks with a higher proportion of households with an annual average income below the county’s median income based on 2013-2018 census data. We will organize our recruitment efforts into phases that will target one to four watersheds at a time. Finally, we will avoid sending recruitment materials to properties that were listed as having destroyed homes in the 2020 Oregon wildfires because it is unlikely these homes would be occupied and any reminders of their homes could be traumatic to survivors. We will conduct engagement activities to increase awareness of the study such as sharing our information at county fairs, farmers markets, on trusted listservs, providing articles for local newspapers and newsletters, and promoting stories about the study on local radio and TV stations.

Targeted mailings in English and Spanish will be sent to the identified households that describe the study, invite them to participate, and receive a free well water test (for arsenic, nitrate, and lead) worth approximately $130 U.S. We will follow up with a postcard two weeks later. Individuals across the state who hear about the study through community engagement activities can add their names to a mailing list and receive recruitment materials.

In order to qualify for the study, participants need to be 21 years or older, be able to complete a questionnaire in English or Spanish, be the homeowner of the property with the well, have owned that property for at least one year and have the intention of owning that property one year after joining the study. After providing informed consent, participants will complete our baseline survey either online through our REDCap portal or by requesting a paper version. After completion of the baseline survey, we will mail them a water test kit that includes written and pictorial instructions to collect their drinking water sample directly from their kitchen faucet. The water kits are designed and prepared by a contracted and certified analytical chemistry testing lab. Participants choose how they would return the water sample (mail via self-addressed prepaid envelope, drop off to local staff, or request staff to come to their home).

De-identified test results will be delivered electronically from the lab to the study environmental health scientist to review and send results letters to participants. The template for the results letter is written in plain talk and was reviewed and approved by the IRB. Participants will be invited to the intervention if the test result indicates that arsenic is above 0.01 milligram per liter (mg/L) and/or nitrate is above 10 mg, or lead is above 0.015 mg/L. These contaminant values were chosen as inclusion criteria because they corresponded to their respective National Primary Drinking Water Regulations maximum contaminant level (54). For those samples with a nitrate level over the MCL and a bottle-fed infant in the home, we established a protocol based on the recommendation of the DSMB to notify and invite these families to navigator trial arm. If multiple members from the household want to participate in the RCT, the participant with oversight or control of their home hazards management (i.e. well, septic) needs to be the primary contact. All household members can participate in sessions.

2.5. Participant assignment into the Be Well Home Health Navigator Program

Upon receipt of test results, eligible households will be assigned by the Program Coordinator into trial arms. We will use an unbiased participation allocation process to balancd socio-economic status across the arms. We will sort households by reported household income level into 3 income strata based on Oregon county U.S. Census median income levels (low-income strata: $0 - $50,999 per year; middle-income strata: $51,999 - $75,999 or households that preferred not to disclose their income; and high-income strata: $76,000 and above.) Visual Basic for Applications (VBA) code randomly sorts participants into one of the two arms within each income strata. Given the rolling recruitment, we will randomize the first household in each income strata to one of the study arms. Then, the subsequent eligible household within the same income category will be allocated to the other study arm. Upon the entry of the next pair of households within an income stratum, a new round of random assignments will be conducted.

2.6. Arm 1 – Usual Care

Participants will receive written materials from Oregon Health Authority’s domestic well water safety program including the Water Well Owner’s Handbook: A guide to water wells in Oregon (655), treatment decision guides developed by our team (56-58), and information about financial incentives for drinking water well improvements. These treatment decision guides include different treatment options with pros and cons plus average one-time costs and maintenance costs. Information about financial incentives for water treatment includes USDA low-interest grant/loan programs. Intervention functions delivered in this arm are education and persuasion (59). These materials are sent in three batches to match the number of in-person intervention contacts in arm 2 to control for attention bias.

2.7. Arm 2 – Community Health Navigator

Participants assigned to this arm will be connected with a well water navigator and receive the usual care described above. The navigator is a CES Community Educator trained in well water safety and motivational interviewing who will provide individualized support for well-owner’s decision-making. Motivational interviewing is a collaborative, goal-oriented style of communication, that is designed to empower people to change by drawing out their own meaning, importance and capacity for change (60). Each household will have up to 3 meetings with their navigator over an approximate 3-month time frame. These meetings occur in person, via teleconferencing, or telephone. During the first meeting, the navigator can provide the following: i) Interpretation and feedback on water test results, ii) Teach-back moments to improve health literacy and numeracy, iii) Assessment of household risk for contaminant exposure from well and from septic (46, 47), iv) Assessment of risk to individual family members, pets, livestock, or gardening, v) Coach to resolve ambivalence using elicit-provide-elicit motivational interviewing processes; vi) Assistance with decision-making and weighing financial options, and vii) Goal-setting and develop action plan (e.g., hold family discussions, review different treatment options, review family finances). The second and third meeting include a follow-up brief motivational interviewing session; pros and cons of their different treatment plans; feedback and refinement of action plan; how to identify or evaluate private remediation services; budget preparation; determine eligibility and connect to any regional government-incentivized home repair program. Intervention functions delivered are education, persuasion, plus personalized training and enablement with the added navigator component (59). These strategies are described in Table 1.

Table 1.

List of intervention strategies used in the two intervention arms of the Be Well Home Health Navigator Program

Arm 1: Usual care	Materials provided
Free water test	After baseline survey
12 Simple Things You Can Do to Protect Your Well	Include in first batch of mailings once in trial arm
Printed Oregon Health Authority Water Well Owner’s Handbook	Include in first batch of mailings once in trial arm
Toll-free number to call with questions	Include in first batch of mailings once in trial arm
Treatment decision guides for arsenic, lead, or nitrate	Include in the second batch of mailings once in trial arm
Information about loan programs and other financial incentives to treat	Include in the third batch of mailings once in trial arm
Arm 2: Access to Cooperative Extension Service (CES) Community Educator for 1-3 individual sessions	Examples of activities during session
All materials provided with usual care (Arm 1)	CES Community Educator provides information about health risks, prevention strategies, financial incentives, and pros and cons of treatment options
Interpret and get feedback on results	CES Community Educator reviews results, when safe to drink water, and answer questions
Improve health literacy and numeracy through teach-back moments	CES Community Educator asks the participant to describe, in their own words, the level of contamination in their well, what puts them at more or less risk, and recommended actions to take. Participant can also explain this information learned to their family members.
Assessment of household risk for contaminant	Participant completes worksheets on drinking well and septic system assessment tools. These worksheets are done in-between meetings and then reviewed with CES Community Educator during next session for feedback on risks.
Assessment of risk to individual family members, pets, etc.	Participant and CES Community Educator discuss the general health risks with this level of exposure for participants, family, and pets..
Coaching to resolve ambivalence or lack of motivation using elicit-provide- elicit motivational interviewing processes	CES Community Educator explores the participant’s understanding and need for information, then provides new information in a neutral manner, followed by exploring what this information might mean for the participant, using a question such as, “What does this mean to you” or “How do you make sense of all this?” Finally, they discuss possible changes or options (for instance, “What have you read about treatments for [contaminant]?.”
Assistance with decision-making and weighing financial options	CES Community Educator review treatment decision guides and asks participant to explain the difference and which treatment work would work best for them and their family.
Goal setting and develop action plan	CES community Educator and participant create a list of small steps and selects one to work on such as talking to partner about finances, contacting treatment companies about price and logistics, listing of pros and cons of each treatment system. CES Community Educator follows up with progress, success, and barriers at next session.

2.8. Intervention training and fidelity

We will assess and monitor fidelity using the framework provided by the NIH’s Behavioral Change Consortium for health behavior change trials (61). To prevent protocol deviations and drift, the study staff will observe 50% of sessions immediately after training and will review 20% of sessions either by recording or in-person throughout the trial. We will assess fidelity of intervention delivery (i.e., what is taught), receipt (i.e., what is learned) and enacted (i.e., what is actually used) using checklists, exit interviews and behavioral strategies (such as set goal to research treatment options, discuss finances with family, completed risk assessments). CES Community Educators complete encounter logs and a checklist of activities following each session. Follow-up surveys with participants confirm which topic or skill they learned and which they performed after the session. Both navigators and participants will provide feedback on number, topics, timing and pacing of the sessions.

2.9. Measurement of effects of the intervention strategies

The reach and implementation of the intervention will be assessed using the Re-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) framework for intervention planning and evaluation (62). Reach will be assessed from the number and sociodemographic characteristics of participants. Implementation will include number and quality of materials developed and direct observation of sub-sample of intervention sessions. Dose will include counts of sessions and number of risk assessments, goal sheets, and action plans developed.

3. Outcomes of the Be Well Home Health Navigator Program

3.1. Logic Model

A logic model describing the activities and outcomes is presented in Figure 2. All participants will complete survey assessments at baseline, 6 months, and 12 months. Well water tests will only be conducted at baseline and 12 months post-baseline to match seasonality of water quality.

Figure 2.

A logic model describing the Be Well Home Health Navigator Program

3.2. Well water analysis

The water test kit includes two trace clean sample bottles (a 125ml bottle preserved with <4C H2SO4 pH<2 for nitrate + nitrite, and a 100ml bottle preserved with HNO3 pH<2 for arsenic and lead) and instructions in English and Spanish for how to collect a water sample from their kitchen faucet. We opted to analyze water samples for nitrate+nitrite (rather than just nitrate) because they can be preserved and kept stable which makes mailing water samples to the lab feasible. The water samples are analyzed by a laboratory that has an up-to-date Oregon Environmental Laboratory Accreditation Program certification using EPA method 200.8 (Determination of Trace Elements in Waters and Wastes by Inductively Coupled Plasma-Mass Spectrometry).

3.3. Water treatment and consumption

Our pre-specified primary outcome is installation of an appropriate treatment device. Appropriate treatment will be defined as the use of drinking water technology that removes the contaminant detected (e.g. reverse osmosis, distillation, arsenic removal filter, or removal of lead-based water fixtures) or switching to bottled water for drinking. At baseline, participants will be asked about what current treatment devices they have installed. Participant will select if treatment system is a point of use under the sink or a whole home system. They will select the specific types of systems installed from the following: pitcher-type water filter, refrigeration filtration system, carbon filter, distillation, reverse osmosis, UV light, anion exchange (resin or iron-oxide filter), sediment filter, water softener. Participants will answer if they only drink bottled water and if they have recently changed their plumbing. Space is provided for additional treatment options as well as why they chose to treat their water this way. Participants will complete a 24-hour and 7-day recall of drinking water (type, amount, frequency) which includes photographs to help estimate water volumes. Finally, participants will be asked about changes made to their water treatment system at 6- and 12-month follow-ups.

3.4. Health history and symptoms

Information about the health of participant and spouse will be collected using the National Health and Nutrition Examination Survey (NHANES) questionnaires specific to hospitalization utilization, cancer, diabetes, cardiovascular disease, blood pressure, anemia, and reproductive health (63). The Patient-Reported Outcomes Measurement Information System (PROMIS) v1.2 Global Health Scale will be used to assess health symptoms (64). Questions about the health of the youngest child under the age of 18 come from the National Survey of Children’s Health and focus on learning and cognitive delays, diagnoses of anxiety, depression, autism spectrum disorder, or attention deficit disorder as well as asthma, respiratory, and gastrointestinal conditions (65).

3.5. Knowledge, Risk Perception, and Attitudes

This study assesses individual-level modifiable outputs known to relate to water testing and treatment behavior (66) including knowledge of contaminants and treatment options, perception of toxicity and susceptibility to risk, attitudes and confidence towards well upkeep. Our baseline survey questions include 1) cognitive risk perceptions, attitudes, norms, self-efficacy and perceived susceptibility using the RANAS scales (67), 2) other motivators and barriers to treatment and testing (such as cost or having a pet or livestock), and 3) environmental health literacy (68). Questions include prior experience with well and septic systems, barriers and use of well water stewardship behaviors adapted from previous studies (23, 26, 69) .

3.6. Individual-level non-modifiable demographics and housing characteristics

Demographic questions are derived from NHANES-based questionnaires for: age, sex, education, income, housing characteristics, and number in household (63).

3.7. Statistical approach to assess program effects

Logistic regression analysis will be used to estimate and compare the proportion of households that implement an appropriate water treatment technology in treatment arms 1 and 2. The primary endpoints are (i) a binary indicator of whether or not an appropriate water treatment was implemented (this could be using bottled water or the appropriate type of treatment) assessed at 6 and 12 months post-baseline and (ii) change in concentration of contaminant (arsenic, nitrate, and lead) assessed 12 months post-baseline. The model will examine the overall effect of the intervention on the change in knowledge, and water treatment and maintenance behavior from baseline to 6- and 12-months follow-ups. The odds ratio will be calculated to measure the strength of the association between the intervention and the change in behavior at each time point. Individual factors will be adjusted for in the model (e.g. age, income, education, full-time employment, years living in the house).

To test the effects of each intervention component, we will replace arm assignment with the behavior change strategies delivered in the intervention (such as goal setting). We will consider the strategy performed if a) participant remembered learning the strategy with the navigator and 2) participant did the strategy at home (such as talked with their family about finances). Each intervention strategy will be added individually to the model to test its direct and indirect mediation effects. Direct mediation effects will be determined through the causal steps according to Baron and Kenny (70): 1) causal relationship between intervention arm and outcome, 2) causal relationship between mediator and intervention arm, and 3) relationship of intervention strategy (i.e. mediator) with outcome when controlling for intervention arm. We can test the use of multiple strategies or the interaction between performing two strategies.

3.8. Sample size justification

Figure 1 displays the overview of the RCT. We anticipate mailing out 4,600 recruitment letters with an approximate 47% response rate to our survey (n=2,200). We estimate 15% of water tests will be contaminated for arsenic, nitrate, or lead (n=330) and these households will be invited to the trial arms. We estimate that 65% will agree to participate in one of our two trial arms (n=215) and that 70% will complete the program and follow-up assessments (n=150).

We calculated the sample size for the repeated MANOVA (multivariate analysis of variance) for two groups and three measurements with two tails, testing power of 0.80 and 0.95, significance level of 0.05, and Cohen’s levels of small (f2 = 0.1), medium (f2 = 0.25), and large (f2 = 0.4) effect sizes. For a small effect size, the required sample size is 967 for a power of 0.80 and 1,545 for a power of 0.95. For a medium effect size, the required sample size is 158 for a power of 0.80 and 248 for a power of 0.95. For a large effect size, the required sample size is 64 for a power of 0.80 and 100 for a power of 0.95. Effect sizes for the relationship between the intervention arm and main outcome are presumed moderate to large and are based on relationships observed between well water testing with action knowledge, self-efficacy and confidence (0.25 < r < 0.35) published elsewhere (26). We used G*Power software to calculate sample size (71) .

3.9. Costs estimates and feasibility of cost utility analysis

Cost and anticipated benefits will be estimated for each intervention arm (Table 2). A Cost-Utility analysis (CUA) is a special use case of cost/effectiveness that weights observable health states by preferences or utility judgments of quality (72) and the benefits of interventions or preventive programs are expressed in terms of quality-adjusted life years (73-75). To estimate quality-adjusted life-years, we will use health outcome data from the PROMIS measures. We will apply a weighting system that uses PROMIS scores to place each individual, at each point in time on a wellness continuum bounded by 1.0 for optimal function and 0 → 0 death (75-77 77). The second component of the method will include microsimulation to estimate expected future health trajectories in the two study arms. The simulation will use the change in daily average intake of arsenic between baseline and end of the follow up period because it has the most robust exposure-response data (18, 78). We will estimate the cost of health conditions reported and/or health conditions that might accrue to participants, their partner, and/or child based on their exposure level in each study arm using data from the Medical Expenditure Panel Survey. A series of sensitivity analyses will evaluate how varying assumptions used in the modeling impacts conclusions.

Table 2.

Lists of variables used to calculate the costs in the Be Well Home Health Navigator Program.

Component	Activities
Participants reach and recruitment	Advertising, social marking direct costs & staff time; recruitment letters; staff time
Connection with stakeholders and advisory board	Travel time & per diem; room rentals for meeting; staff and stakeholder time
Well water tests at baseline and follow-up	Costs to conduct the baseline & follow-up water tests; costs to print & mail test results
Arm 1: Written materials and usual care	Costs to print well owner handbooks and contaminant guides; adapt website as needed for this specific project. Costs to print study materials, costs to upload to website; Staff time staff time to mail books and handle incoming calls related to the study
Arm 2: Written materials and access to a navigator	Creation of training materials for environmental health navigators; Staff effort to attend training on intervention procedures, to perform quality control checks, and to attend annual booster training sessions. Staff effort to contact, schedule, meet and deliver 3 sessions with participants and to complete forms and write-up notes for next meeting. Mileage for in-person and long-distance costs for telephone intervention sessions; costs to print study materials.

4. Potential for program sustainment and dissemination after study completion.

As the numbers of states implementing well water safety legislation increase (879), it is critical to offer programs to aid well owners with adherence to these new policies. If successful, our safe drinking water intervention could be immediately scalable to other counties and states in the U.S. through existing communication channels among Land Grant Universities and their respective Cooperative Extension Services.

Highlights:

• Approximately 25% of private wells in the United States that are used for drinking water contain contaminants that are known to increase the risk of chronic diseases.

• Homeowners are responsible for maintaining the quality of the drinking water provided by their private wells.

• This intervention leverages the trusted relationships of Cooperative Extension Service Community Educators within local communities to educate, assist, and motivate well owners to mitigate arsenic, nitrate, or lead in their well water that is found to be above safe drinking water standards.