Introduction The completion of the Human Genome

The completion of the Human Genome Project in 2003 opened many doors for

scientists and health professionals to identify and potentially prevent common disorders

through gene analysis. Over the last decade, this progress has not only impacted clinical

medicine and individual patients but has also shown the ability to reduce morbidity and

mortality of susceptible populations through more personalized public health

programming (Auffray et al., 2016; Cragun, Lewis, Camperlengo, & Pal, 2016).

Unfortunately, translation and implementation of genomic advances has been slow, both

in clinical medicine and public health (National Academies of Sciences, Engineering,

2016). Most research in this area has been focused on implementation in the clinic;

however, evidence supports the use of genomic technology for population-wide chronic

disease prevention. Therefore, gaining an understanding of the opportunities and

challenges to public health genomic implementation is prudent at this time.

In this study, I conducted a quantitative survey of state and territorial chronic

disease directors (CDDs) in the United States to examine what genomic activities are

currently being achieved and determine if there is an association between these state

activities and what these CDDs know or are interested about in chronic disease genomics.

This study was important at this time because of recent evidence-based recommendations

for screening at-risk individuals and their family members for hereditary forms of three

chronic diseases; breast cancer, colon cancer, and cardiovascular disease. Identification of

this group of individuals and subsequent treatment or preventative measures could reduce

the morbidity and mortality from these conditions and allow for positive social change

through increased health and quality of life for those affected.

This introduction to the study will include background information leading to the

current landscape, the purpose of this study and why it is important, and a description of

the problem. I will also provide an explanation of the research questions and hypothesis, a

description and justification of the theoretical framework, and outline the nature and

significance of the study. Finally, I will define terms specific to this study and clarify the

study’s assumptions, scope, delimitations, and limitations.

Background

The Healthy People 2020 initiative has included genomic activities for the first

time, signifying increasing evidence that family history and genetic testing can be used to

promote health benefits in clinical and public health capacities (Valdez, Yoon, Qureshi,

Green, & Khoury, 2010; Weir et al., 2015). State public health genomics activities have

traditionally been focused on newborn screening (NBS); however, evidence and test

availability has prompted recommendations for adult population screening initiatives

(Green, Dotson, Bowen, Kolor, & Khoury, 2015). The expansion of public health

genomics from newborn screening into chronic disease is important and timely

considering the impact of new molecular technology and research advances in the field

(Bowen, Kolor, Dotson, Ned, & Khoury, 2012). Pilot public health genomics programs in

chronic disease have showed that advances can be made by conducting evaluations and

examinations that support use of genomic information and family history for disease

prevention efforts; integrating this information into existing programs; and developing

and circulating educational materials for health care providers, policy makers, and the

public (St Pierre et al., 2014).

The last time a survey of state genetic activities was performed was in 2001

(Coalition of State Genetics Coordinators, 2002; Piper et al., 2001). Also at that time,

Kaye et al. (2001) made very specific recommendations regarding the need for the

integration of genetics into public health and how genomic activities were connected to

the core functions of assessment, assurance, and policy development. These authors also

provided the rationale for and details of responsibilities for a state genetics coordinator

position in order manage activities and facilitate collaborations in genomics. Another

analysis of the role of genetics in the provision of essential public health functions found

that these programs provide for many public health obligations including diagnosing and

investigating health problems and hazards in the community (NBS), mobilizing

community partnerships with genetics professionals and other health care providers, and

linking the population to needed personal health services (Wang & Watts, 2007).

Although these studies support the use of genomics as a public health tool outside

of NBS, implementation beyond the pilot programs has been slow. In 2010, the

Association of State and Territorial Health Officials (ASTHO) published a State Public

Health Genomics Resource Guide outlining novel approaches for public health

departments to translate genomic science into public health practice using examples from

a limited collection of states with innovative programs (ASTHO, 2010). Although this

toolbox was created to help other states find ways to integrate genomics into public

health, the slow adoption requires investigation into why genomic services for chronic

diseases at the state level is not supported. A 2006 survey of state health officers

confirmed important emerging public health functions; however, genomics was not one of

them (Beitsch, Brooks, Grigg, & Menachemi, 2006). Moreover, a 2011 survey of chronic

disease public health professionals about their training needs did not include any

questions about genomics (Wilcox, Majestic, Ayele, Strasser, & Weaver, 2014). Until

public health practitioners begin to think about genomics as a viable tool for public health

prevention, implementation of state programs will likely not become a priority. This study

was needed at this time to assess the current status of knowledge and interests in

genomics by state CDDs and identify opportunities and challenges to increasing

awareness of the importance of genomics by this group in light of the new chronic

disease genomic testing recommendations.

Problem Statement

The Centers for Disease Control and Prevention (CDC) Office of Public Health

Genomics (OPHG), the U.S. Preventive Services Task Force (USPSTF), and others have

evaluated evidence and formulated recommendations for hereditary forms of chronic

disease conditions that would benefit from patient genetic counseling, testing, and

cascade screening of family members (Dotson et al., 2014). These applications are

divided into a three-tier classification system with Tier 1 genomic applications having

clear evidence for practical implementation (Khoury, Coates, & Evans, 2010). Initial

Tier1 applications have been identified as hereditary breast and ovarian cancer (HBOC),

Lynch syndrome (LS), and familial hypercholesterolemia (FH; Bowen et al., 2012).

Despite recommendations and evidence to support screening for these conditions, only a

limited number of states are working in this area (Green et al., 2015).

Federal funding from the CDC OPHG and the Division of Cancer Prevention and

Control (DCPC) to support state chronic disease genomics infrastructure development,

surveillance activities, and implementation of evidence-based recommendations has been

limited to a small number of states since 2003: Michigan, Minnesota, Oregon, Utah,

Connecticut, Colorado, Georgia, and Ohio (CDC, 2016; Green et al., 2015; St Pierre et

al., 2014). Furthermore, to date, most implementation strategies have been focused

specifically to address initiatives in HBOC, less often on LS, and limited activity on FH.

((Laufman, Duquette, & Trepanier, 2012; Nordestgaard et al., 2013; Trivers, Rodriguez,

Cox, Crane, & Duquette, 2015). Other states, such as Washington, Hawaii, Illinois, and

New York, have made incremental strides in state genomics planning without CDC

funding ((ASTHO, 2010; Trivers et al., 2015)).

The ability to utilize these evidence-based initiatives under the current climate

could be problematic and negatively impact our public’s health if citizens have limited

access to these screening programs. Although studies have shown that chronic disease

departments are hindered by poor collaborations, shifting goals, lack of organizational

support, limited resources, alternating priorities, and competency by the public health

workforce (Allen et al., 2013; Alongi, 2015), understanding and leadership by CDDs

could also impact program implementation. Examining this group to evaluate their

knowledge and interests in genomics is important in order to assess their ability to engage

in these new technologies and support promotion of funding opportunities. Additionally,

establishing an association between certain current state genomic activities and the

knowledge and interests of CDDs could provide the evidence needed for increased

coordination and funding to the states.

Purpose of the Study

The purpose of this quantitative survey design was to determine whether there is

an association between current state genomics funding or specific state genomic activities

and the level of knowledge and interests in genomics by state CDDs. In this quantitative

survey design study, I analyzed the results of a survey of CDDs in all U.S. states and

territories. My intent with this study was to identify and describe particular activities or

particular states that may be associated with an increased level of knowledge and interest

in genomics by CDDs and which may also influence implementation of Tier 1 genetic

tests at the state level.

Research Questions and Hypothesis

Research Question 1: To what extent, if any, is there an association between states

that have received funding for chronic disease genomics and the level of knowledge and

interests in genomics by state and territorial CDDs?

H01: There is no association between states that have received funding for chronic

disease genomics and the level of knowledge and interests in genomics by state and

territorial CDDs.

H11: There is an association between states that have received funding for chronic

disease genomics and the level of knowledge and interests in genomics by state and

territorial CDDs.

Research Question 2: To what extent, if any, is there an association between

current state genomic activities and the level of knowledge and interests in genomics by

state and territorial CDDs?

H02: There is no association between any current state genomic activities and the

level of knowledge and interests in genomics by state and territorial CDDs.

H12: There is an association between one or more current state genomic activities

and the level of knowledge and interests in genomics by state and territorial CDDs.

Current state genomic activities that were seen as having a potential impact on

chronic disease public health genomics program implementation were queried. These

included (a) a state genetics needs assessment, (b) a state genetics needs assessment that

includes chronic disease conditions, (c) inclusion of genetics in the state public health

action plan, (d) genetic educational programs, (e) genomics topics on the BRFSS, (f)

analysis of state cancer registries or other vital records data to identify citizens with

hereditary cancer syndromes, (g) frequency of collaborations or partnerships with outside

entities related to genomics, and (h) presence of legislation and/or regulation specifically

related to genomics. To determine the knowledge and interests in genomics by the CDDs,

they were asked about their (a) awareness of contact information for clinical genetic

services for potential referral or consultation, (b) knowledge of Tier 1 recommended

conditions, (c) agreement with genomic statements, and (d) interest in the integration of

genomic activities.

Theoretical Framework

In this study, I used the theoretical framework of the diffusion of innovations

theory to explore the adoption of chronic disease genomics at the state level. According to

Rogers (2003), adoption of new innovations in organizations can be challenging even if

the advancements have clear, evidence-based rewards as in the current climate of Tier 1

recommendations. Diffusion is a process that occurs over time through communication

between members of a social system and culminates with a modification of the structure

and function of the social system (Rogers, 2003). The four main elements of this theory

include the characteristics of the innovation itself, communication channels, time, and a

social system that supports adoption (Rogers, 2003)

In this study, I examined the social system that supports adoption of chronic

disease genomics, particularly looking at specific state genomic activities that may be

associated with the level of knowledge and interests in genomics by CDDs. Moreover,

based on Roger’s (2003) theory, adoption of chronic disease genomics is ready to move

into more states at this time; I will discuss this topic in more detail in Chapter 2. This

adoption could be connected to genomic champions who have worked to secure funding

for genomic activities in their states. My determination of whether or not there was an

association of greater knowledge and interests in genomics by the CDDs in the few states

that have received funding could influence the need to identify a genomic champion in

each state and secure more funding for state genomic activities.

Nature of the Study

In this study, I used a nonexperimental, cross-sectional, correlational quantitative

survey design to examine current state genomic activities and possible associations with

the level of knowledge and interests in genomics by CDDs from all U.S. states and

territories. This survey was originally developed by a subcommittee of the American

Public Health Association (APHA) Genomics Forum Policy Committee (GFPC), of

which I am a member, in order to determine opportunities and challenges of state CDDs

in genomics and possibly create a position statement by the APHA. The survey design

was chosen because it would be fairly easy to administer to the study group via e-mail

web-link, be simple to develop at little or no cost, and could ask a number of pertinent

questions to obtain a broad range of data (see Frankfort-Nachmias & Nachmias, 2008).

With the first half of the survey, I collected information about the presence of

specific state genomic activities and in the second half inquired about knowledge and

interests in genomic topics among CDDs. State genomic activities were either present

(Yes), absent (No), or unknown (Don’t know). One question was framed to inquire about

frequency of collaborations with outside entities and was measured on a Likert scale.

Knowledge and interests about genomic topics were also measured on a Likert scale;

however, these results were converted to a numerical product for analysis of the level of

knowledge or interest (e.g. Agreement with genomic statements: 1= strongly disagree

through 5= strongly agree). I performed statistical analysis to determine the relationship

between the variables to see if state funding or any particular activity was associated with

an increased level of knowledge and interests in genomics by the CDDs.

Definition of Terms

I am providing the following definitions to ensure uniformity and understanding

of these terms throughout the study:

Cascade screening: The systematic identification and testing of family members

of an individual who has a particular disease of interest (Ned & Sijbrands, 2011).

Diffusion: The process by which an innovation is communicated through certain

channels over time among members of a social system (Rogers, 2003).

Evidence-based medicine: Health technologies and practices supported by sound

research evidence (Greenhalgh, Robert, Macfarlane, Bate, & Kyriakidou, 2004)

Genetics: The study of heredity with a focus on a specific and limited number of

genes with known function in disease (Manolio, 2016).

Genomics: The study of an individual’s entire genetic makeup, the genome, while

also examining how the genome interacts with environmental or behavioral factors. This

is especially important in the study of complex chronic diseases that affect large factions

of the population (Cragun et al., 2016)

Innovation: A novel set of behaviors, routines, and ways of working, which are

directed at improving health outcomes, administrative efficiency, cost-effectiveness, or

user experience, and which are implemented by means of planned or coordinated action

(Greenhalgh, Robert, & Bate, 2005).

Precision medicine: Tailoring medical therapies to subcategories of disease based

on genomics (Ashley, 2015).

Precision public health: Providing the right intervention to the right population at

the right time (Khoury, Iademarco, & Riley, 2016).

Public health genomics: The study and application of knowledge about the

elements of the human genome and its functions, including interactions with the

environment, in relation to health and disease in populations (Cleeren, Van der Heyden,

Brand, & Van Oyen, 2011).

Scope and Delimitations

The scope of this study consisted of all state and territorial CDDs in the United

States. To date, this group had not been surveyed specifically about knowledge in

genomics and its connection to chronic disease nor the possible association with current

state genomic activities. Each state and territory has one director and all are members of

the National Association of Chronic Disease Directors (NACDD), which was my point of

contact and access to this study group. The total possible participants for this study was

58, and each chronic disease director had an equal chance to participate in this study. To

encourage participation, I limited the quantitative survey design to mostly closed-ended

questions and took place in a 6-week timeframe.

I explored a range of theoretical models to provide a framework for this study.

Consideration was given to the transformational leadership and transtheoretical models,

utilization management, attribution, and complex adaptive theories. Upon discussion with

my committee, I decided to use Rogers’ diffusion of innovations theory to explain the

current state of chronic disease public health genomics.

Assumptions

One assumption I made in this study was that the database provided for the target

population by the NACDD was current and accurate. The CEO of the NACDD explained

that the database is updated yearly and frequent e-mails are sent each month ensuring

accuracy (J. Robitscher, personal communication, January 5, 2017). Another assumption

was that the sample of survey responses is representative of the whole population of

CDDs being studied. Because one question on the survey asked about what state the

chronic disease director practices in, this helped determine if the study population resided

in different parts of the country (heterogeneous sample) or from states that have received

some type of funding for state genomic activities related to chronic disease. A final

assumption was that respondents answered the questions truthfully. As this is a

confidential, voluntary questionnaire, it would be more likely that these directors would

be honest in their responses.

Limitations

Due to the small sample size available for the study, results may not be

generalizable beyond the specific population from which the sample was drawn. Because

the number of CDDs in the United States and territories is limited, a small response rate

impacted the power of the analysis (< 80%) by introduction of Type II errors and not

allowing for generalizability to the study population (Field, 2013). The nonresponse bias

is also important; the nonresponse bias is how different or similar those who do not

respond are from the whole survey population (Johnson & Wislar, 2012). Those who

chose to complete the survey may have had more knowledge and interest about genomics

and vice versa. Also, because the subcommittee and I conducted this study voluntarily,

there was no monetary compensation for the directors to complete the survey, which may

have also impacted the response rate (Cho, Johnson, & VanGeest, 2013). This survey was

a web-based instrument with a short completion time (6 weeks), which could have

affected participation if the CDDs did not find the time or remember to complete it in the

allotted window. A final limitation was that, although this was a confidential web-based

survey, this did not assure that the chronic disease director who received the survey was

the one who completed it in part or in totality.

Significance

Data from this study could provide public health leaders at the federal, state, and

local levels with information as to what specific genomic activities are associated with

increased levels of knowledge and interest in genomics by state CDDs. This information

could be starting point for states to increase genomic activities to conform to the new

Healthy People 2020 objectives, Precision Medicine Initiative, and Tier 1 genetic testing

and screening recommendations (Auffray et al., 2016; Dotson et al., 2014; Weir et al.,

2015). These results could also support and provide rationale for public and private

funding for state genomic activities and identify states that are ready to begin genomics

implementation. Finally, the results of this survey could provide a small snapshot of what

is and is not being done in the states today in regards to genomics as well as highlight the

level of knowledge and interest in specific genomic topics by CDDs.

Implications for Social Change

Walden University defines positive social change as “a deliberate process of

creating and applying ideas, strategies, and actions to promote worth, dignity, and

development of individuals, communities, organizations, institutions, cultures, and

societies” (Walden University, 2016, para. 2). This definition is expected to provide a

foundation for student research that will result in the betterment of human and social

conditions. In the context of this study, the implications for social change relate to the

rights of the public to have access to evidence-based technology that has the ability to

reduce morbidity and mortality of certain hereditary diseases. Gaining an understanding

of the opportunities and barriers to Tier 1 chronic disease genomics implementation could

provide the “ideas, strategies, and actions” for increasing this work in more states

(Walden University, 2016, para. 2). If there is a connection between particular state

genomic activities and how knowledgeable and interested the CDDs are in this area, this

finding could encourage funding for more state activities. This funding could provide a

ripple effect of engagement for chronic disease departments, state health departments, and

other stakeholders.

Summary

Individuals who work in public health have a responsibility to ensure that the

communities they serve are healthy by assuring safe, accurate, and accessible chronic

disease genomic services (Cragun et al., 2016). Expansion of state chronic disease

genomics to a larger proportion of states is prudent at this time in light of the Healthy

People 2020 objectives, the Precision Medicine Initiative, and evidence-based Tier 1

genetic testing and screening recommendations (Auffray et al., 2016; Modell, Greendale,

Citrin, & Kardia, 2016; Weir et al., 2015). The purpose of this study was to identify some

possible opportunities and challenges to reach that goal. This foundation will support

ideas to facilitate implementation of chronic disease public health genomics in more

states.

In this introduction, I presented the statement of the problem, research questions,

definition of terms, scope, nature, and significance of the study, as well as study

limitations. In Chapter 2, I will provide a review of the literature in the field of public

health genomics from its inception to the present day and connect what is currently

happening through Roger’s diffusion of innovations theory. I will outline the

methodology and data collection procedures in Chapter 3 and describe the data analysis

and findings in Chapter 4. Finally, Chapter 5 will include my summary and discussion of

the findings along with conclusions and recommendations for future studies.

Chapter 2: Literature Review

Introduction

The problem I addressed with this study was the need for public health genomic

programming to move beyond traditional NBS and into the chronic disease arena in the

United States and its territories (Bowen et al., 2012). Current evidence is available to

encourage genomic risk assessment through the screening and testing of individuals and

cascade follow-up and testing of at-risk family members for hereditary forms of breast,

ovarian, and colon cancer as well as cardiovascular disease (Dotson et al., 2014). Public

health practitioners, especially at the state level, are poised to be the leaders in facilitating

evidence-based genomic surveillance and screening for certain hereditary chronic

diseases (Green et al., 2015).

This chapter will include a narrative of the birth of public health genomics

through the present day including how public health genomics ties into the core public

health functions, the role of genomics in population health, a definition of precision

public health, Tier 1 genetic testing recommendations, and the burden of each condition.

Finally, I will describe the role of state chronic disease departments in Tier 1

implementation and the current landscape. Through my review and discussion of the

literature on the framework of this study, Rogers’ diffusion of innovations theory, I will

demonstrate where public health genomics is today and how it is poised to move to the

next level, what translation barriers exist, and how collaboration is crucial to

implementation success.

The purpose of this quantitative survey design was to determine whether there is

an association between current state genomics funding or specific state genomic activities

and the level of knowledge and interests in genomics by state CDDs. Using this

quantitative survey design, I analyzed results of a survey of CDDs in all U.S. states and

territories. The intent of this study was to identify and describe particular activities or

particular states that may be associated with an increased level of knowledge and interest

in genomics by CDDs, which may also influence implementation of Tier 1 genetic tests

at the state level.

Literature Search Strategy

I conducted a review of the literature using the Walden University Library,

Google Scholar, and the World Wide Web. I searched the CINAHL,

MEDLINE/PubMed, Science Direct, EBSCO, and ProQuest Central databases for

peerreviewed, English language journal articles focusing on articles of interest in the last

5 to 6 years (2010–2016). Key search terms used included public health, genetics or

genomics, chronic disease directors, knowledge of genomics, population genomics,

champion, and diffusion of innovations. These searches yielded 326 scholarly journal

articles, four books, and two dissertations related to the topic of this study.

Theoretical Foundation

At the time of this study, very few states are doing any significant work in chronic

disease public health genomics with a majority of states incorporating these activities on

a limited basis and predominantly focusing on the core function of assurance (Laufman et

al., 2012). Those few states who are leading the implementation of Tier 1 testing play an

important role in the diffusion of these new evidence-based public health applications by

modeling activities, providing public health outcome data, and championing the cause as

an opinion leader or change agent. In this literature review, I investigated how Rogers’

diffusion of innovations theory could be used to explain the development of the field of

chronic disease public health genomics and ways that this theory could be used with the

results of this study to identify clear avenues to increase adaptation across more states

throughout the country.

Rogers’s Theory

Diffusion is “the process by which an innovation is communicated through certain

channels over time among members of a social system” (Rogers, 2003, p. 5). The four

main components are (1) the innovation, (2) communication channels, (3) time, and (4)

the social system. An innovation is a new idea that will likely bring forth a certain degree

of uncertainty to the social system depending on the number of alternatives available and

the probability that the new innovation is superior to or enhances current practices

(Rogers, 2003). Communication of the benefits of the new innovation is a two-way

process that occurs over many cycles of information exchange to reach a mutual

understanding (Rogers, 2003). Time is an important component of the diffusion process

whether to understand the innovation decision process from first knowledge to

acceptance or rejection, why certain individuals adopt the innovation earlier or later in the

process, and the innovation’s rate of adoption in a system (Rogers, 2003). Finally,

diffusion has the capability of altering the structure or function of a social system when a

new idea is conceived, diffused, and either incorporated or rejected leading to

consequences that change the social system (Rogers, 2003).

The five adopter categories are (1) innovators, (2) early adopters, (3) early

majority, (4) late majority, and (5) laggards (Rogers, 2003). Innovators, the first to adopt

a new idea into a system, actively seek new ideas, often reach outside their own locale for

information and support, and are able to handle greater amounts of uncertainty (Rogers,

2003). These individuals often serve as change agents who influence others in the social

system to adopt the innovation (Rogers, 2003).

Early adopters are also enthusiastic about new innovations while still being

selective about what they adopt (Rogers, 2003). This group also has a great degree of

opinion leadership; however, they examine both the positive and negative aspects of an

innovation, so buy-in from this group is critical to adoption success (Rogers, 2003).

These individuals are often consulted by others for advice and information about the

innovation, serve as role models, and help decrease uncertainty by others (Rogers, 2003).

The early majority adopts new innovations before the average members of the system;

however, adoption comes only after lengthy deliberation (Rogers, 2003). This group is an

important link to the diffusion process by connecting the enthusiastic leaders with the

typical members of the group who are inclined to be more resistant to adoption (Rogers,

2003).

The late majority are skeptical and only adopt innovations due to economic or

system pressures even after they have been persuaded of the utility of the new idea; most

if not all of the uncertainty of adoption must be removed before this group will join

(Rogers, 2003). Finally, laggards are the last to adopt a new innovation because they

traditionally live in the past, are resistant to change, and are suspicious of new ideas

(Rogers, 2003). Rogers (2003) explained that by the time laggards adopt a new

innovation, it may already be out of date and surpassed by a newer method.

How fast an innovation is adopted by the members of a social system is

contingent on its perceived characteristics, type of innovation decision and

communication channels used, inherent nature of the social system, and the extent of

effort by change agents in diffusing the innovation (Rogers, 2003). The diffusion effect

explains the important relationship between the rate of knowledge about an innovation

and the rate of its adoption by those in the system (Rogers, 2003). As the level of

innovation knowledge increases to the 20–30% range, only small amounts of adoption

occur; however, once this threshold passes the tipping point (which can be slightly

different depending on the innovation and system), the rate of adoption increases

exponentially (Griliches, 1957). This threshold is often contingent on the point at which

opinion leaders in a system begin to look favorably on the innovation and activate peer

networks in the social system.

How Roger’s Theory Has Been Used in Previous Research

According to Schon (1963), resistance to change is normal and may also seem

desirable in many instances to assure stability in organizations. In order to promote

changes that are in the best interests of the organization and those they serve, a champion

will often emerge to fight for the introduction and development of a new innovation.

These champions are most often emergent leaders from within the organization who are

effective at influencing the leadership process to produce the desired change (Taylor,

Cocklin, Brown, & Wilson-Evered, 2011). Champions are intrinsically motivated,

energetic and enthusiastic, and committed to the cause; either the new advancement finds

a champion or dies (Schon, 1963).

Previous researchers have examined the use of champions on diffusion of

innovations in health care settings. A 2006 study on the implementation of the MOVE!

weight-management program in the Veteran’s Health Administration found that

organizational readiness for change and the presence of an innovation champion were key

factors in the success of this program (Weiner, Haynes-Maslow, Kahwati, Kinsinger, &

Campbell, 2011). Novick et al. (2015) also found that champions who advocated for the

enactment of a new model for prenatal care were instrumental in successful

implementation and sustainability at group practices. Finally, a study about the adoption

of the Agency of Healthcare Research Quality tools to assess pharmacy’s health literacy

practices also found that a change champion would have a positive impact (Shoemaker,

Staub-DeLong, Wasserman, & Spranca, 2013).

Roger’s Theory and Public Health Genomics

The field of public health genomics outside of NBS has seen some great successes

in model states since the establishment of the Office of Public Health Genomics at the

CDC in 1997, the completion of the Human Genome Project in 2003, and the

formalization of the field in 2005 (Green et al., 2015; Modell et al., 2016). These model

states have all received some kind of funding for genomic activities and assessments

(ASTHO, 2010; CDC, 2016; Green et al., 2015; St Pierre et al., 2014; Trivers et al.,

2015) Table 1 shows the model states and their category of adoption. At this time, the

field appears to be at or over the tipping point and is ready for integration of chronic

disease genomics into more states (19–20) in the early majority category.

Table 1

Current Status of Chronic Disease Public Health Genomics Programming Adoption

Adoption category

% expected

N based on 58 states

and territories

States involved

Innovators

2.5%

1–2

Michigana,b,c

Oregona,b,c

Early Adopters

13.5%

7–8

Connecticuta,b,c

Utaha,b,c

Minnesotaa,b

Georgiaa,b

Coloradoc,d

Ohioa

Washingtone

Early Majority

34%

19–20

Late Majority

34%

19–20

Laggards

16%

9–10

Note. (a) St Pierre et al., 2014, (b) Green et al., 2015 (c) CDC, 2016, (d) ASTHO, 2011a,

(e) ASTHO, 2011b.

Innovativeness refers to how quickly or reluctantly an individual or system unit

adopts an innovation (Rogers, 2003). New innovations are often proposed by opinion

leaders, such as the OPHG, who maintain a high degree of credibility regarding the

technical and theoretical aspects of the innovation (Rogers, 2003). Change agents, who

are often in the early adopter category, work alongside opinion leaders to champion the

adoption of an innovation (Rogers, 2003). In this case, the change agents would

champion the diffusion of chronic disease genomics at the state level.

One area that has been shown to impact an innovation’s adoption is the

communication and influence that occurs through social networks, peer and expert

opinion, champions, and change agents (Greenhalgh et al., 2004). Currently, model states

that are doing work in chronic disease public health genomics have at least one individual

who devotes time to initiate agendas, develop and assess programs, seek and obtain

funding, provide education, and facilitate stakeholder collaborations (ASTHO, 2010; St

Pierre et al., 2014). As Rogers (2003) explained, the “presence of an innovation champion

contributes to the success of an innovation in an organization” through communication of

the benefits of an innovation over a period of time and numerous conversations to

influence the rate of adoption (p. 414). Schon (1963) clearly stated that without a

champion, new ideas will likely die from normal and somewhat desirable resistance to

change. Having a genomic champion at the state level that is knowledgeable and

committed to chronic disease genomics could positively impact the rate of adoption of

genomics in state health departments.

Although evidence exists to support chronic disease genomics in the states, only a

handful of states are working to that end. Diffusion of this new innovation and the rate of

adoption by more states can be encouraged by a variety of factors such as the level of

communication and influence provided to stakeholders, which includes state and

territorial CDDs. Innovator and early adopter states have at least one genomics champion

who has likely impacted the rate of adoption in their states and this can be a model to

increase implementation in additional states.

History of Public Heath Genomics

In 1990, the National Institutes of Health and Department of Energy launched the

Human Genome Project to develop technology that could analyze DNA, map and

sequence the human genome, and investigate associated ethical, legal, and social issues

(National Human Genome Research Institute (NHGRI), 2015). In order to address the

population health impact of the HGP, the CDC established the Office of Genetics and

Disease Prevention (now the OPHG) and created a strategic plan to address the

translation of genomic advances into population health (Zimmern & Khoury, 2012).

Since its inception in 1997, the OPHG has involved many partners to anticipate, evaluate,

and demonstrate the translation of genomics into population health practices (OPHG,

2011). A meeting in Bellagio, Italy in 2005 resulted in the formal definition of public

health genomics as “a multidisciplinary field concerned with the responsible and effective

translation of genome-based knowledge and technologies into health care practices to

improve population health” (Bellagio Report, 2005; CDC, 2007, p. 1). Public health

genomics seeks to use population data of genetic variation and environmental influences

to establish evidence-based interventions for disease prevention and health improvement.

Use of genome-based knowledge for public health interventions has been around

long before the term public health genomics was first defined. In 1961, Dr. Robert

Guthrie developed the first NBS test to identify infants with phenylketonuria resulting in

the institution of mandatory state screening programs in 1963 (NHGRI, 2016). Today,

virtually all babies born in the United States undergo NBS in state run programs to detect a

variety of endocrine, metabolic, and hematologic conditions that are genetic in nature and

was named one of the 10 Great Public Health Achievements of the 20th century (CDC,

2011; Ross, 2010). Universal screening of newborns highlights the ability of public health

to reduce morbidity and mortality of hereditary conditions through state-run programs.

Recently, there has been increasing momentum from the national level to

encourage the integration of genomics into public health programming. In December of

2010, the Office of Disease Prevention and Health Promotion included genomics in the

Healthy People 2020 objectives for the first time (Office of Disease Prevention and

Health Promotion, 2016). These new objectives reflect increasing evidence to support the

use of genetic tests and family health history in clinical medicine and public health. The

first two recommendations include (a) Women with a high familial risk of breast, ovarian,

tubal, or peritoneal cancer could benefit from genetic counseling to learn more about

genetic testing for the breast cancer susceptibility gene (BRCA) 1/2 mutations and post-

test surgical options to reduce risk, and (b) All newly-diagnosed colorectal cancer patients

should receive information regarding genetic testing to identify a hereditary form of this

cancer (LS), which could benefit family members by reducing their risk of colorectal

cancer caused by LS through screening and interventions.

In January 2015, President Obama announced his support for the Precision

Medicine Initiative, which aims to link researchers, providers, and patients to focus

disease prevention and treatments based on individual differences in genetics,

environment, and lifestyle (The White House, 2015). His $215 million, 2016 budget

financing is to be a collaborative public and private investment in genomic advances,

tools for managing and analyzing large sets of data while protecting patient privacy, and

health information technology. This initiative is also designed to engage at least a million

Americans to volunteer their health information to study health outcomes, develop new

treatments, and introduce a more precise and personalized healthcare system. Of course,

all of these promises are at-risk under the new administration.

How Public Health Genomics Fulfills the Core Public Health Functions

The mission of public health is “fulfilling society’s interest in assuring conditions

in which people can be healthy” (Institute of Medicine Committe for the Study of the

Future of Public Health, 1988, p. 7). This mission is to be carried out through public and

private partnerships, however, public agencies have a responsibility to assure that

essential components are in place to address the mission effectively. Along with that

mission are three core public health functions; assessment which includes collection and

analysis of population health data, assurance of quality services to all, and policy

development based on sound scientific knowledge and use of the democratic political

process. Table 2 provides some examples of genomics in relation to these core public

health functions.

Table 2

Role of Genomics in the Delivery of Essential Public Health Services

Core Function

Description

Examples

Assurance

Link to /

Provide Care

Assure A

Competent

Workforce

Evaluate

•

Collaborate with other public and private

entities and educate public health staff and

private health-care workers about the use of

genetic information to improve healtha

Incorporate genomics into the curricula of

medical schools, nursing schools, and

schools of public health and provide

opportunities for continuing education

around genomicsb

•

Evaluate genomic tests, services, and

information to ensure availability, efficacy,

accessibility, safety, quality, and ethical

practices while also enforcing the policies

and standards enacted to ensure thisb

•

Identify and analyze the factors that

influence the impact of genetic information

and the delivery, utilization and quality of

genetic tests and servicesa

Assessment

Monitor Health

Diagnose &

Investigate

•

Utilize family history or genetic testing to

identify at-risk individualsb

Perform epidemiologic studies on the

prevalence of genetic risks factors /

variants within the community to

determine their contribution to identified

health problemsa

•

Study gene-gene and gene- environment

interactionb

•

Assess the availability, appropriateness,

and accessibility of quality genetics

resources in the communitya

•

Assess the impact of genetic information

and its value in improving healtha

•

Research the community’s and health care

providers’ knowledge of the use of

genetics to improve healthc

(table continues)

Core Function

Description

Examples

Policy Development

Inform, •

Educate, &

Empower

Mobilize

Community

Partnerships

•

Develop

Policies

Enforce Laws

Improve genomic literacy of the public,

health care practitioners, policy makers, and

other stakeholders through audiencespecific

educational initiatives about the integration

of genomics into health promotion and

disease prevention programsb

In collaboration with stakeholders,

implement regulatory policies and

guidelines for clinical applications, test

implementation, use, impact, and protection

of genomic information, and accessibility

and quality of genomic technologya,b

•

Identify and analyze the economic, social,

ethical and political implications of

advances in human genetics, including the

information and communications needs of

stakeholdersb

•

Assure insurance coverage for high risk

individualsb

•

Develop, enhance and sustain partnerships

with key partnersa

Note: (a) Khoury, 2011, (b) McWalter & Gaviglio, 2015, (c) ASTHO, 2010.

The Role of Genomics in Population Heath

During the last 20 years, the OPHG at the CDC and other multidisciplinary groups

have been trying to use the knowledge gained from the HGP and other scientific advances

and translate this into activities for population health (Zimmern & Khoury, 2012).

Beyond the ever-expanding state universal newborn screening panels, public health

genomics is going to play a significant role in epidemiological studies, infectious disease,

chronic disease, and environmental health (Roberts, Dolinoy, & Tarini, 2014).

Implications will also be felt in areas such as biostatistics, health policy and regulation,

health education, health behavior responses to genomic information, and equitable

distribution of the costs and benefits from genomic discoveries and applications. The

integration of genomics into this wide variety of activities will require complex

structures, processes, and collaborations by a diverse range of stakeholders to fully realize

the translation of genomic findings into improved population health.

Family Health History

The original genomic tool used in medicine and public health has been the use of

family health history (FHH). A large majority of chronic diseases of public health

significance including diabetes, cardiovascular disease, several cancers, osteoporosis, and

asthma have been shown to have a strong family history component (Yoon et al., 2002).

FHH is a combination of shared genetic susceptibility, environment, and behaviors and,

prior to the availability of genetic testing, chronic disease programs and clinical health

practitioners have traditionally focused their efforts on the environmental and behavioral

components (OPHG, 2011). Use of FHH and possible genetic screening and testing could

complete a three-legged stool of disease prevention targets. Public health leaders can be

effective advocates in educating others about the link between FHH and chronic disease,

especially to minority groups, and can use this tool as a surveillance method to identify

at-risk individuals and their family members, and evaluate the impact on population

interventions (Butty et al., 2012; Khoury et al., 2011; Powell, Edleson, O’Leary,

Christianson, & Henrich, 2011; Senier et al., 2015). Although FHH will continue to be a

valuable primary prevention tool, issues with collection, standardization, interpretation,

and integration with electronic health records exist and will need to be addressed (Bowen

et al., 2012; Modell, Kardia, & Citrin, 2014; Valdez et al., 2010; Williams, 2012).

From Precision Medicine to Precision Public Health

Precision medicine, sometimes referred to as personalized medicine, is a concept

that implies prevention or treatments based on individual differences (Collins & Varmus,

2015). Pharmacogenomics, the most recognized precision medicine mechanism, is aimed

at providing the right drug to the right patient at the right time based on an individual

patient’s genetic makeup (Auffray et al., 2016). Precision public health, on the other

hand, focuses on individuals within a defined population for “providing the right

intervention to the right population at the right time” (Khoury, Iademarco, & Riley, 2016,

p. 398). Long before advances in genomics, Rose (1985) explained that populations

would be healthier and costs contained if prevention efforts were targeted at those in the

population who are at greatest risk for an identified disease. Identifying and explaining

why some individuals, or groups of individuals, get sick while others don’t is an excellent

guide to public health prevention efforts. Targeted public health screening programs not

only seek to protect susceptible individuals, but try to discover and control the cause of

incidence; susceptibility will not exist if causes are removed or circumvented.

The completion of the Human Genome Project has enabled a significant

opportunity to practice clinical medicine and public health in a novel way. Acquiring the

ability to identify disease in individuals and populations based on genetic components

will permit us to target prevention efforts and treatments based on heredity and individual

or population susceptibility. This has been highlighted by inclusion of genomics for the

first time in the Healthy People 2020 objectives and support for the Precision Medicine

Initiative. The field of public health genomics fulfills the core public health functions,

supports delivery of essential health services, and will play an active role in furthering

use of genomic advancements in population health.

Tier 1 Genetic Testing Recommendations

Due to advances from the Human Genome Project and genomic applications,

evidence-based recommendations are now available to move public health genomics from

reducing morbidity, mortality, and disability in the newborn period to identification of

genetic influences across the lifespan (Bowen et al., 2012). Although a large proportion of

applications will be delivered in the clinical care setting, state public health agencies are

poised to be the leaders in targeted population screening programs (Khoury et al., 2011).

These leaders will be responsible for program development and implementation, delivery,

assessment, reduction of potential harms, equitable access, and creation of a

multidisciplinary infrastructure for program support and future use of genomic

applications. Public health professionals, who are focused on population health and

reduction of health disparities, can successfully gather these stakeholders without bias or

another agenda.

Tier Classifications

In 2005, the CDC OPHG established the Evaluation of Genomic Applications in

Practice and Prevention (EGAPP) Initiative to develop a process for an evidence-based

evaluation of genomic tests in clinical medicine and public health practice (Teutsch et al.,

2009). Recommendations for test readiness come from a multidisciplinary expert group

of nonfederal, independent individuals who evaluate a test’s (a) analytic validity (ability

to accurately and reliably measure the genotype of interest); (b) clinical validity (ability

to detect or predict the disorder or phenotype of interest); (c) clinical utility (evidence of

improved measurable clinical outcomes and usefulness to patient management); and (d)

associated ethical, legal, and social implications before suggesting its use (Green et al.,

2015; Secretary’s Advisory Committe on Genetic Testing, 2000). These tests have been

classified into a three tiered, color-coded system to indicate what tests are ready to be

integrated into clinical care and public health practice (Dotson et al., 2014; see Table 3).

Table 3

Tier Classification System of Genomic Tests

Tier 1

Tier 2

Tier 3

Genomic and Family

Health History

applications have a base of

synthesized evidence to

support integration into

practice.

Genomic and Family

Health History applications

have insufficient evidence to

support routine

implementation into

practice.

These applications do have

the potential to provide

information for informed

decision making by

patients and providers or

for informing selective use

strategies (e.g. clinical

trails) through clinical or

public health policy

decision making.

Genomics and Family

Health History applications

have evidence that either

results in recommendations

against use OR no relevant

evidence is available at this

time. Tier 3 applications are

not ready for routine use but

may be used for clinical or

population research.

Note. Adapted from Dotson, W. D., Douglas, M. P., Kolor, K., Stewart, a C., Bowen, M.

S., Gwinn, M., … Khoury, M. J. (2014). Prioritizing genomic applications for action by

level of evidence: A horizon-scanning method. Clinical Pharmacology and Therapeutics,

95(4), 394–402. http://doi.org/10.1038/clpt.2013.226.

As of April 2016, 46 tests have been classified as Tier 1, 105 are Tier 2, and nine

are in Tier 3 (OPHG, 2016c). Although most of the Tier 1 genomic tests are related to

pharmacogenomics, tests also include the 31 core newborn screening conditions and three

chronic disease conditions. Applications for these conditions include cascade DNA and

LDL (low-density lipoprotein) testing of relatives of patients identified with FH,

diagnostic screening for LS for colorectal cancer patients and cascade screening for their

family members, and risk prediction and referral to genetic counseling for BRCA testing

for those with a risk of HBOC.

Classifications for these Tier 1 genomic applications are based on the

recommendations of the National Institute for Health Care Excellence, EGAPP, and the

U.S. Preventive Services Task Force (Evaluations of Genomic Applications in Practice

and Prevention Working Group, 2009; National Institute of Health and Care Excellence,

2008; U.S. Preventative Services Task Force, 2014). All three disorders are autosomal

dominant (only one defective gene is required to inherit the disease); have lifelong health

implications; and use family health history to identify those at risk and cascade screening

to connect to family members who could benefit from further follow-up (Bowen et al.,

2012). FH also includes a rarer homozygous variant (two mutations are inherited) which

increases the severity of the disorder (Nordestgaard et al., 2013).

Public Health Burden of Tier 1 Conditions

There are over two million individuals who carry the mutational genes for HBOC,

LS, and FH in the United States today (George, Kovak, & Cox, 2015). Considering that

cancer and heart disease are currently the top two burdens in our health care system,

costing an estimated $75 and $320 billion respectively per year, finding ways to identify

and reduce the incidence of these diseases should be a key public health focus (American

Cancer Society, 2016; American Heart Association, 2015). At present, these conditions

are poorly identified by the healthcare system so targeted Tier 1 genetic testing programs

and potential cascade screening of family members implemented through state public

health departments in collaboration with health care practitioners could offer significant

reduction in health risks from these diseases and their associated costs (OPHG, 2014).

Table 4 displays the potential impact of Tier 1 genetic testing on the conditions identified.

Table 4

Public Health Prevalence and Burden of Tier 1 Conditions

Condition

Breast cancer

Colon

Ovarian cancer cancer

Cardiovascular

disease

3rd most

Estimated total new

adult cases 2016

Prevalence of

mutation in new

cases

(table continues)

Conditi

Breast

cancer

Ovarian

cancer

Colo

cance

Cardiovascul

disease

Potential

identification

per

year through

Tier 1

recommendatio

nsb

Risk of disease

with

mutation

4932–

17,26

perso

2,228–

3,342

persons

2858–

4763

perso

N/A

40–

80%3

11–40%3

80%

50% men

–

30%

women7

1:300–

1:500 3*

1:300–1:500

1:370

1:250c–

1:5007

Most common

in women1

2nd highest

cause of cancer

death1

5% of cancer

deaths in

women1

common in

men and

women1

3rd highest

cause of

death1

Leading cause

of death in both

men and

women

246,6601

women

2,600 men1

22,2801

95,2701

73.5 million

in pop (31.4%)

2–7% 2a

10–15%2a

3–5%4

N/A

Estimated

prevalence

mutation in

population

Estimated

US population

carrying

mutationd

Yearly deaths

648,000

–

1,080,0

648,000–

1,080,000

875,67

600,000–

1.2

million8

40,450

women

440

men1

14,4201

49,190

370,0009

Note: (a) Significantly higher rates in Ashkenazi Jewish population, (b) Based on calculation of total new

cases and % new mutation prevalence. Does not include family members potentially identified through

cascade screening, (c) Estimated higher rates in European Caucasian populations, (d) Based on estimated

current U.S. population of 324 million (United States Census Bureau, 2016b).

1. (American Cancer Society, 2016), 2. (D’Andrea et al., 2016), 3. (Petrucelli, Daly, & Feldman, 2013),

4. (American Cancer Society, 2014), 5. (Hampel & De La Chapelle, 2011), 6. (Guillén-Ponce,

MolinaGarrido, & Carrato, 2012), 7. (National Organization for Rare Disorders, 2016), 8. (Ned &

Sijbrands, 2011), 11. (American Heart Association, 2015)

Description of Tier 1 Conditions

Hereditary breast and ovarian cancer (HBOC). HBOC syndrome is caused by

mutations in the BRCA1 and BRCA2 genes, which not only cause an increased risk for

breast (40–80%) and ovarian (11–40%) cancers, but also pancreatic and prostate cancers

(Petrucelli et al., 2013). Mutations in BRCA 1/2 are passed in an autosomal dominant

fashion and cluster in families (U.S. Preventative Services Task Force, 2014). Prognosis

of these cancers is based on when the cancer is detected, so identification of increased

susceptibility through genetic testing and subsequent preventative monitoring and/or

prophylactic surgery could impact the morbidity and mortality triggered by these

mutations. For women with these mutations, surgery could reduce the risk of breast and

ovarian cancer by 69% (U.S. Preventative Services Task Force, 2014).

Current USPSTF recommendations are to screen women for a strong family

history of increased risk for harmful BRCA 1/2 mutations and those identified should be

offered genetic counseling and potential BRCA 1/2 mutation testing. USPSTF gives this

recommendation a ‘B’ rating which means that they recommend provision of this service

because there is a high certainty that the net benefit of this service is moderate and a

moderate certainty that the net benefit is moderate to substantial (U.S. Preventative

Services Task Force, 2016). This ‘B’ rating is significant because this allows coverage by

the Affordable Care Act (Henry J. Kaiser Foundation, 2016). Beyond the USPSTF

recommendations, some authors propose that population screening for the BRCA 1/2

mutation would be cost-effective in the high-risk Ashkenazi Jewish population

(D’Andrea et al., 2016).

Lynch syndrome (LS). LS, also called hereditary nonpolyposis colorectal cancer

(HNPCC), is the most common cause of hereditary colon cancer accounting for 3–5% of

all colorectal cancers (American Cancer Society, 2014; Guillén-Ponce et al., 2012). These

individuals and their families are also at greater risk for other cancers including

endometrial, ovarian, and stomach (Guillén-Ponce et al., 2012). LS is an autosomal

dominant cancer syndrome caused by mutations in DNA mismatch repair (MMR) genes

MLH1, MSH2, MSH6, and PMS2, which function to correct mismatched base pairs as

well as tiny insertions and deletions that occur during DNA replication (Mange et al.,

2015). These corrections are needed to decrease genomic instability which occurs during

DNA synthesis and mutations in these genes will lead to rapid tumor growth

(GuillénPonce et al., 2012). Evidence also shows involvement by an epithelial cell

adhesion molecule (EPCAM), which indirectly affects DNA repair by causing the MSH2

gene to be turned off (Kempers et al., 2011).

Screening for LS consists of tumor testing by either immunohistochemistry (IHC)

or microsatellite instability (MSI) followed by genetic sequencing and deletion analysis

of the MMR genes depending on the results of IHC and MSI (Mange et al., 2015). A

hallmark of LS is the early-onset of colon cancer diagnosis (< 45 years), so prompt

identification of patients, tumor testing, and cascade screening of their family members

could lead to a reduction in LS-caused colorectal cancer incidence and related mortality.

At this time, evidence supports the integration of Tier 1 condition identification

and prevention into clinical and public health practice. Tier 1 conditions have been shown

to have genomic and family health history validation to support inclusion of Tier 1

genetic testing and screening for susceptible individuals and their family members. Tier 1

conditions include HBOC, LS, and FH, which are all contributing to the large

prevalence and burden of cancer and heart disease experienced in the United States today.

Familial hypercholesterolemia (FH). FH is an autosomal co-dominant disorder

expressed with abnormally high concentrations of low-density lipoprotein cholesterol

(LDL-C), which increases an individuals’ risk of premature coronary heart disease (CHD)

from atherosclerosis (Ned & Sijbrands, 2011). FH is caused by loss-of-function mutations

in the LDL receptor (LDLR) and apolipoprotein (APOB) genes and gain-offunction

mutations in the proprotein convertase-subtil-sin/kexin type 9 (PCSK9) gene (Austin,

Hutter, Zimmern, & Humphries, 2004). Untreated, FH poses an approximate overall 20-

fold increase in CHD from the general population, which includes a 50% CHD risk in

men by age 50 and a 30% risk of CHD in women by age 60 (Marks, Thorogood, Neil, &

Humphries, 2016).

Elevated LDL-C levels in affected individuals begin even before birth and those

with two abnormal genes (FH homozygotes) can develop CHD very early in life and die

before age 20 if left untreated (Nordestgaard et al., 2013). FH is as common as Type I

diabetes, and more common than cystic fibrosis or Down’s syndrome, however, it is

estimated that only 1-25% of all cases are diagnosed (Knowles et al., 2014; Modell et al.,

2016; Ned & Sijbrands, 2011; Nordestgaard et al., 2013). Identification of individuals at

risk for FH includes high levels of cholesterol and family and/or personal history of early

onset CHD. Targeted screening of these individuals for the FH mutation, subsequent

treatment with lipid lowering pharmaceuticals, and a program of diet and exercise could

lead to the prevention of tens of thousands of heart attacks over these individual’s

lifetimes.

Implementation of Tier 1 Tests: Current Landscape In 2002, a

Chronic Disease Directors’ Summit was convened to begin the dialogue and develop a

plan to move genomics out of NBS and into chronic disease units (Association of State

and Territiorial Health Officials, 2002). These experts asked the CDC to help states

respond to the resulting information and applications from the Human

Genome Project. In 2003, the CDC funded four states, Michigan, Minnesota, Oregon, and

Utah in 5-year cooperative agreements to integrate genomics into their state public health

programs (ASTHO, 2010; Green et al., 2015). With annual awards of between

$150,000$250,000, these states were able to use internal and external planning to

integrate FHH and genetic testing results into existing genetics and chronic disease

policies and programs (St Pierre et al., 2014). In addition, they formed partnerships,

evaluated public data, developed workforce capacity and leadership, and established

justifiable interventions using FHH, assessments, and educational curricula. After the

infrastructure was built in the first round, the CDC issued new 3-year cooperative

agreements to Michigan and Oregon in 2008 in order to shift the focus from capacity

building to translational activities in public health genomics (ASTHO, 2010). The

program focus for these agreements was on HBOC surveillance, education, and policy

development in support of the USPSTF 2005 recommendations.

In 2011, the OPHG shifted state funding of genomic activities to the Division of

Cancer Prevention and Control (DCPC) at the CDC to highlight the focus on cancer

genomics by the Healthy People 2020 objectives. The DCPC granted $300,000 per year

from 2011–2014 to Michigan, Oregon, and Georgia for HBOC activities and by the end

of the three-year period, each had an established, effectively operating breast cancer

genomics program (Trivers et al., 2015). Recently, Utah, Connecticut, and Colorado have

been included in the support of implementation of evidence-based cancer genomics

recommendations (OPHG, 2016a). Also in 2011, the Connecticut Department of Public

Health Genomics Branch received a Healthy People 2020 Action Award to facilitate

cancer genomics in that state (Connecticut Department of Public Health, 2012). Other

states (Ohio, Hawaii, Illinois, and Washington) have fostered state genomic activities

through implementation grants, academic centers, and collaborations with outside

stakeholders, but none have been funded to the extent that these model programs have

(Green et al., 2015).

Examples of Cancer Genomics Translation by Model States

The following list describes some ways that model states have integrated cancer

genomics into their public health programming:

• Addition of breast cancer questions to the Behavioral Risk Factor Surveillance

System (BRFSS).

• Analysis of state cancer registry data and using this information to communicate

and educate providers and patients about potential HBOC counseling and testing

of patients and their family members.

• Development of new surveillance systems with key stakeholders and genetics

clinics to evaluate uptake of HBOC genetic counseling, testing and follow-up.

• Collaboration with public health clinics to integrate HBOC risk screening into the

clinical intake process.

• Analysis of and collaboration with insurance companies to assure coverage for

genetic counseling and testing for HBOC.

• Education of health providers and the public about Tier 1 tests.

(Trivers et al., 2015)

National Resources for State Public Health Genomics

In 2010, ASTHO developed a State Public Health Genomics Resource Guide

highlighting the issues, strategies, and challenges to state genomic implementation with

links to appropriate resources (ASTHO, 2010). This publication also described specific

activities and tactics that model states were accomplishing to meet clearly defined

genomic objectives. In 2014, the OPHG published an online Genomic Application Toolkit

to share the public health genomics methods of these model programs and give other

states some ideas and advice for development and application of genomic programs in

their own states (OPHG, 2016a, 2016b). The website explains what the Tier 1 genomic

applications are, their importance to population health, and how state and local health

departments can play an important role in the application of Tier 1 tests by identifying

people who could benefit from testing and extending that benefit to their family members.

This toolkit also provides links to implementation videos and other resources to help.

The Role of State Chronic Disease Departments in Tier 1 Implementation

Because the current Tier 1 genetic testing recommendations outside of NBS and

pharmacogenomics are all identified as hereditary forms of chronic diseases, state and

territorial CDDs and the personnel who work in these departments should be educated

about and engaged in Tier 1 genetic testing program implementation (Zimmern &

Khoury, 2012). Currently, most genomic expertise in state and territorial health

departments falls within maternal and child health as it relates to NBS issues, however,

there now needs to be a greater understanding in chronic disease departments about the

impact of genetics on population health and how to use evidence-based recommendations

to implement new practices. Unfortunately, many chronic disease programs across the

country are being reduced, are often underfunded, and are not standardized or as

comprehensive as they need to be, especially as they relate to genomics (Allen et al.,

2013; Maylahn, Fleming, & Birkhead, 2013). Moreover, it is agreed that a majority of

public health professionals have not been educated adequately on genomics, whether

through public health education or on-the-job training (Department of Health and Human

Services, 2011; Marzuillo et al., 2014). It will be important for CDDs to play a leadership

role in the integration of genomics into state chronic disease plans, assessment of

program effectiveness, education of their workforce and the communities they serve, and

initiation and facilitation of collaborations with stakeholders.

By 2010, all states had a Comprehensive Cancer Control (CCC) plan in place and

one study found a significant increase in genomics-related terms in these plans from

2005-2010 (Laufman et al., 2012). These CCC plans included goals and strategies related

to FHH, public and provider awareness of genetics and genomics (education), breast

cancer referrals, access to genetic services, support and expansion of partnerships,

development and promotion of screening (diabetes), and increased research funding

(Alzheimer’s disease). The increase in genomic activities could have been in response to

the Healthy People 2020 goals and/or evidence-based recommendations for cancer

genomics; however, this study found that genomics still hasn’t grown as a priority at most

state levels (Laufman et al., 2012). Although evidence-based public health genomic

strategies have been shown to have a significant impact on population health, the

commitment for genomics at the state level has been slow due to a shortage of

organizational leadership and support, lack of understanding, and limited resources for

competing priorities (Allen et al., 2013)

Currently there a small number of model states doing work in the area of chronic

disease public health genomics and most are accomplishing their goals in small

increments with minimal funding. That being said, the role of state public health

departments, specifically chronic disease units, in light of the Healthy People 2020

objectives, Precision Medicine Initiative, and Tier 1 recommendations can and should

increase to meet these imperatives. Providing evidence of successful programming and

studying models that work could help obtain funding from sources who may benefit from

such integration.

Challenges and Opportunities

Implementation Barriers

Since the completion of the Human Genome Project in 2003, the fields of genetics

and genomics have developed rapidly, yet translation “from bench to bedside” and more

so from “bench to community” has been a slow process (Cornel, Van El, & Borry, 2014).

Studies show that the translational process, from research evidence to clinical practice, is

17 years, however, only 14% of all discoveries actually make it there (Khoury et al.,

2007). Calculating from the date of gene discovery of the Tier 1 tests, FH (1985), LS

(1987), and HBOC (1995), places the translational timeline to clinical and public health

application as 2002, 2004, and 2012 respectively (Brown & Goldstein, 1986; Krainer et

al., 1997; Lynch et al., 2009; Morris, Wooding, & Grant, 2011). As was discussed

previously, clinical practice guidelines and public health recommendations have been in

place for some time, yet clinical medicine and public health practitioners are slow to

adopt suggested practices.

An expected reason for slow adoption is funding. Ninety-eight percent of genomic

funding is in the research discovery phase and “bench to bedside” applications while <

2% is devoted to population translation and outcomes research (Khoury, Gwinn, Bowen,

& Dotson, 2012; Laurence, 2012). Lack of evidence and data showing health outcomes

makes it hard to advocate for genomics program funding in the states and outcome data

from model states is limited. Ironically, governmental support is often based on

translational research data and translational research cannot be accomplished without

governmental support (Modell et al., 2014). Moreover, many state chronic disease

programs have been reduced including a 57% reduction in state funding by the CDC from

2013–2014 (Allen et al., 2013; Khoury et al., 2011; Maylahn et al., 2013). These limited

resources along with a lack of organizational support and leadership due to competing

priorities and an opinion that genomics is a low-yield investment compared to current

practices will certainly slow or hinder state adoption rates (Allen et al., 2013; Khoury et

al., 2007). Without strong health data analysis to encourage these evidence-based

intervention (EBI) applications, it will be difficult to encourage states to move beyond

their current chronic disease practices.

Other barriers to genomic translation in state chronic disease departments is the

lack of awareness and education by public health practitioners, health care providers, and

the public (American Public Health Association, 2013; Williams, 2012). Many of the

model states have included genomics educational assessments and/or programs to

increase knowledge and awareness about genomics and recommended health practices (St

Pierre et al., 2014; Trivers et al., 2015). Reimbursement for Tier 1 recommendations and

other genetic services also limits what states are willing to support (Williams, 2012).

Cancer genomics is not a mandated public health program like NBS and, without a

nationalized healthcare system, reimbursement will be dependent on the patient and their

insurance availability (which may vary widely or depend on the recommendation level)

(Bowen et al., 2012). Finally, the limitations of our current electronic health records

system to collect, analyze, and store the large volumes of data could impact health

outcomes data (Williams, 2012). As this is one of the goals of the Precision Medicine

Initiative, this data should be easier to ascertain once a better system is in place.

The Need for Collaboration

Implementation of public health genomics is difficult and the need for

collaboration within state public health departments and external stakeholders is the key

to success (Genetic Alliance, 2014; OPHG, 2011). State public health departments are in

a unique position to foster these collaborations and mobilize partnerships that will ensure

a competent public health and clinical medicine workforce as well as assure accessible

and quality genetic services (Cragun et al., 2016). Because public health departments are

also the only ones who have the legal authority to collect population data in some

jurisdictions, they will need to lead the accumulation of this information and tabulate

health outcome figures (Maylahn et al., 2013). The public health, clinical medicine, and

genetic service providers’ connection needs to be especially strong to assure individual

patients and susceptible populations have the information about and access to pertinent

genetic services. Other important external stakeholders reside in hospitals, academic

centers, local public health organizations, and advocacy organizations (Laufman et al.,

2012). Advocacy organizations are especially important because they help to establish

buy-in from the public and other organizations (Modell et al., 2016).

Some states have a state genetics coordinator who’s scope of practice is beyond

NBS (Coalition of State Genetics Coordinators, 2007). Success in novel public health

interventions implementation has been tied to strong leadership and champions who are

passionate about the program (Milat, Bauman, & Redman, 2015). All of the model states

who have accomplished strides in public health genomics have someone who works part

or full-time in that capacity at the state level or at an associated academic university. This

individual can and should be the leader of the coordination, collaboration, and

communication of state public health genomic services. In addition, collaboration should

occur between public health agencies at the national, state, and local level as well as

regional collaboratives to share ideas and challenges (Alexander, Keehn, Kaye, &

O’Leary, 2016; Bowen et al., 2012).

Translating genomic advances from the discovery phase to population

implementation and subsequent improvement in health outcomes is not a small task.

Many barriers exist including funding, awareness, knowledge, competing priorities,

reimbursement, and lack of organizational support. In order to overcome these barriers,

state public health professionals need to collaborate with other stakeholders to assure

these advances are accessible to the populations they serve.

Summary and Conclusions

With fully validated and clinical practice guidelines for screening of Tier 1 tests in

place, implementation of screening programs for individuals and their family members at

risk for HBOC, LS, and FH is ready to be launched through collaborations by public

health agencies, clinical medicine practitioners, and advocacy groups (Modell et al.,

2016). Because prevention of population morbidity and mortality is a key public health

endeavor, Tier 1 genetic testing and cascade screening of family members illustrates how

family health history can be modifiable. State chronic disease departments will be

instrumental in the delivery of these programs through the formation of strong

partnerships with many different sectors of the communities they serve.

CDDs are poised to be the leaders of the dissemination and coordination of these

new health promotion practices while assuring a focus on the needs of underserved

populations (American Public Health Association, 2013; Senier et al., 2015). Because of

this, they will need to understand and help facilitate implementation of Tier 1 genetic

testing recommendations. Rogers’s diffusion of innovations theory explains that

communication and influence are impactful to the rate of adoption; encouraging states to

have an individual who can be the champion for chronic disease genomics would help

more states adopt this new innovation.

In Chapter 3, I will outline the methods that were used to perform this study. This

will include a description of the instrument, participants that were studied, and procedures

for data collection, coding, and analysis. The analysis will be specifically explained

through a detailed presentation of the specific variables and statistical tests to be used.

Chapter 3: Research Methods

Introduction

Following Tier 1 genetic screening and testing recommendations, leaders in state

and territorial health departments will be called upon to coordinate, collaborate, and

communicate these initiatives in their areas (Green et al., 2015). This is especially true for

CDDs, who oversee the areas touched by these recommendations for breast cancer, colon

cancer, and cardiovascular health. The main purpose of this study was to determine

whether certain state genomic activities or current genomics funding is associated with

the level of knowledge and interests in genomics by state and territorial CDDs. This could

lead to identification of state activities which may help CDDs to be more informed and

prepared to lead Tier 1 testing and screening in their states. These results may also shed

light on present knowledge and interests by CDDs providing opportunities and insight on

challenges for Tier 1 program implementation in the states and a baseline for future

research. The purpose of this chapter will be for me to describe the research questions and

hypotheses, survey instrument used, participants, study variables, procedures for data

collection and coding, and specific data analysis techniques. Finally, I will also define

potential threats to the validity of the study.

Research Design and Rationale

In this quantitative research study, I employed a nonexperimental, cross-sectional,

correlational survey design to investigate whether there is an association between current

state funding in genomics and/or specific state genomic activities and the level of

knowledge and interests in genomics by current CDDs in all U.S. states and territories.

This study was initiated by a subcommittee of the APHA GFPC, of which I am a member,

in order to determine the current status of state CDDs in regards to readiness for

implementation of Tier 1 testing and identify opportunities and challenges to that end.

The committee agreed that I could perform a secondary analysis of the results of this

survey to answer the research questions about the possible connection between state

genomics activities and CDDs knowledge and interests in genomics.

The survey committee also determined that a quantitative, web-based,

selfadministered questionnaire format would be a quick, efficient, and cost-effective

design to obtain the information sought (see Frankfort-Nachmias & Nachmias, 2008).

Using a quantitative, cross-sectional survey would allow me to collect important

information from the study group at a single point in time and delivery via a free web-

based service that would eliminate costs. The other committee members and I were

conducting this study voluntarily with no outside funding. Because this is a new

instrument developed by the GFPC subcommittee members, there was no known

reliability or validity at the time of this study. Content validity was determined based on

the expertise of the committee members and other experts in areas which were thought to

impact state Tier 1 implementation. A pilot study of the survey was not performed before

delivery to the CDDs.

Quantitative surveys have been used in the past to identify issues in public health

services delivery (Jacobs, Dodson, Baker, Deshpande, & Brownson, 2010; Stamatakis et

al., 2012; Wilcox et al., 2014). Survey response rates have also been studied showing that

multiple methods (mail, Internet, telephone), incentives, and follow-up attempts can

impact response rate (Cho et al., 2013; Dillman, 2015; Millar & Dillman, 2011; Pit,

Hansen, & Ewald, 2013). Moreover, the advent of personal hand-held devices also seems

to be having a negative impact on the completion of surveys (Stern, Bilgen, & Dillman,

2014). Because this was a project taken on voluntarily by the GFPC subcommittee (four

members with no financial support), there was no availability for incentives, mailings, or

telephone calls. The subcommittee decided to deliver the survey via e-mail link to the

Internet with a 6-week timeline during which two follow up e-mail requests would be

made. The benefit of using the Internet includes lower costs, decreased time, and easier

data entry and analysis (Ahern, 2005).

The Chronic Disease Directors Survey consisted of 16 dichotomous, Likert scale,

limited contingency, and demographic questions to determine age, educational degree,

and state or territory (see Appendix). Some questions included a response of “Don’t

know” if respondents were unable to accurately answer the question. The first half of the

survey (nine total questions) contained inquiries about the extent of each state’s activities

in genomics. The second half of the questionnaire was comprised of questions to gauge

the participants’ knowledge and interests of genomic topics specifically as they relate to

chronic disease (four total questions, 18 different topics). Three of these questions were

based on a Likert scale and were, therefore, used to determine the level of knowledge and

interests in genomics (1 (very poor) – 5 (very good)) rating of knowledge of Tier 1

conditions; 1(strongly disagree) – 5(strongly agree) of agreement with genomic

statements and interest in genomic topics.

Methodology

Population

The selected population for a given methodology is the “aggregate of all cases that

conform to some designated set of specifications” (Maruyama & Ryan, 2014, p. 231). For

this study, participants were CDDs from all U.S. states, territories (Guam, Marshall

Islands, Micronesia, Palau, Puerto Rico, Samoa, and U.S. Virgin Islands), and the District

of Columbia (N = 58). This was a select group with a defined number of participants. I

was provided with e-mail contact information for the directors and delivery of the survey

link by the NACDD located in Atlanta, GA (NACDD, 2016b). Founded in 1988, the

NACDD is a nonprofit, public health organization dedicated to supporting CDDs in each

state and territory by connecting over 6,000 chronic disease practitioners to create

partnerships, develop policies, implement programs, and share knowledge about chronic

disease prevention and health promotion (NACDD, 2016a).

Sampling and Sampling Procedures

In this study, I employed a simple random convenience sampling of all CDDs in

the United States and territories (N = 58), and each director had an equal chance to

complete the survey. The survey was disseminated to these directors through their

employee e-mail. I conducted a G*power analysis for the sample size using the t-test

difference between two dependent means (matched pairs), two tailed, with a medium

effect size (0.2), 0.05 a, and 0.80 power, which yielded a sample size of 199. As this

was much greater than the total sample of the population, sample size analysis could not

be used in this case (see Faul, Erdfelder, Buchner, & Lang, 2009).

Procedures for Recruitment, Participation, and Data Collection

In this study, I used data that was originally collected from a quantitative survey

instrument delivered via the Qualtrics survey platform to the membership of the NACDD

(N = 58) from February 11, 2016 through March 31, 2016(Qualtrics, 2015). Each

participant had an equal chance to voluntarily complete the survey with no monetary or

other compensation provided for doing so. One of the GFPC members provided access to

the Qualtrics platform through the University of Michigan Medical School Information

Services with a specific survey link for respondents to connect to the survey

(https://umichumhs.ut1.qualtrics.com/jfe/form/SV_d6IJGrKYOzZZDb7). The other

committee members and I obtained access to this dataset from this committee member.

The CEO of NACDD and the NACDD policy chair were our points of contact in

the organization, authors of the cover letter, and distributors of the survey link via e-mail.

The cover letter explained the purpose and importance of the survey to chronic disease

public health genomics, encouraged participation, and assured confidentiality of

individual results (see Frankfort-Nachmias & Nachmias, 2008). This method was

employed for two reminder e-mails (sent on February 25 and March 21, 2016) during the

survey period.

Instrumentation and Operationalization of Constructs

The Qualtrics survey platform allowed for coding of the survey questions within

the program, automatically assigning a quantitative answer choice value to each selection

within a question (first answer choice = 1, second choice = 2, etc.). The program also

allowed for variable naming and assignment of a label for each question. In order to

measure the variables, I named variables by the question number (Q1, Q2, etc.) and

identified those with subtopics (e.g. Q9_1, Q9_2, etc.). All of this information was easily

downloaded into SPSS for analysis. When this survey was originally entered into the

Qualtrics system, each question was given a number; however, after committee input,

some questions were relocated on the survey and the numbers were not reordered in the

Qualtrics system. Therefore, the final survey questions did not follow in numerical order,

but were identified correctly in SPSS. The Appendix shows the final survey and coding

scheme that I used.

In this survey, I assigned the responses for state genomic activities a number (1 =

Yes, 2 = No, 3 = Don’t know) for all questions except for the frequency of collaborations,

which was based on a Likert scale (1 = In the past quarter, 2 = In the past year, 3 =

Rarely, 4 = Never, 5 = Never but potentially in the future). Additionally, question

subtopics were also coded 1–5 (Q8), 1–7 (Q9), 1–4 (Q13), and 1–9 (Q25) to correspond

to each subtopic from top to bottom. Except for the question regarding awareness of

contact information for clinical genetic services (Yes/No), a Likert scaling method was

used to identify the varying levels of knowledge and interest in genomics by CDDs on a

5-point continuum (knowledge of Tier 1 conditions rated 1–5 (Q10); agreement with

genomic statements (Q11) or interest in genomic topics (Q12) rated 1 = Strongly

disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree, 5 = Strongly agree,

6=Don’t know (Q11) or 6 = We already do this (Q12). Each Likert scale response was

given a quantitative number to correspond with the result and this had a direct correlation

to the level of knowledge and interest in genomics of each survey participant; the higher

the number, the greater level of knowledge and interest and vice versa. Responses of

“Don’t know” were identified and removed as outliers during initial statistical analysis

and then run again to include these results to examine their potential impact.

The purpose of this study was to determine possible associations between state

genomic funding as well as specific state genomic activities and the level of knowledge

and interests in genomics by state and territorial CDDs. Because of this, I considered each

variable potentially dependent on the other. In other words, no variable was considered as

an independent or causative variable. All questions in this survey were categorical

(nominal); however, Likert scaled questions (Q22, Q23, Q24, and Q25) were converted to

continuous values (1–5 or 6) for statistical analysis to allow for determination of a level

of knowledge, interest, or frequency of collaboration by each respondent and the study

group as a whole; the higher the number, the greater the level of knowledge and interest

except for levels of collaboration, which employed a reverse numbering scheme (see

Appendix).

I considered multiple choice questions with a response of “Yes” (1) present and

responses of “No” (2) or “Don’t know” (3) were considered absent initially. “Don’t

know” responses for state activities were changed to (2) to analyze them as not present.

These results were then returned to their original states to analyze the impact of the

“Don’t know” responses. Finally, the “Don’t know” response for Q23 (agreement with

genomic statements) was removed (* in dataset), so it would not be analyzed and impact

the mean value of interest by the CDDs. Tables 5 and 6 explain the variable types and

which statistical tests that I used for each association. Finally, I tested data validity and

reliability with exploratory analysis and Cronbach’s alpha.

Research Question 1 Variables

I identified the following variables and subvariables for Research Question 1:

1. State funding for chronic disease genomics

2. Level of knowledge and interests in genomics by CDDs

Subvariables:

2a. Awareness of contact information for clinical genetic services

2b. Knowledge of Tier 1 recommended conditions

2c. Agreement with genomic statements

2d. Interest in genomic activities

Table 5

Operationalization of Variables for Research Question 1

Variable pair

Question type

Variables included

State genomic funding

Categorical

Identified states with funding

Knowledge and interest

Categorical

Continuous

2b, 2c, 2d

Research Question 2 Variables

I identified the following variables and subvariables for Research Questions 2:

1. State genomic activities.

Subvariables:

1a. State genetics needs assessment

1b. State genetics needs assessment includes chronic disease conditions

1c. Genetics in state action plan

1d. Genetics education

1e. Genomics in BRFSS

1f. Analysis of state cancer registries

1g. Genetic legislation

1h. Frequency of collaborations

2. Level of knowledge and interests in genomics by CDDs

Subvariables:

2a. Awareness of contact information for clinical genetic services

2b. Knowledge of Tier 1 recommended conditions

2c. Agreement with genomic statements

2d. Interest in genomic activities

Table 6

Operationalization of Variables for Research Question 2

Variable pairs

Question type

Variables included

State genomic activities

Knowledge and interest

Categorical

1-2, 1a-2a, 1b-2a, 1c-2a, 1d-2a,

1e-2a, 1f-2a, 1g-2a

State genomic activities

Knowledge and interest

Categorical

Continuous

1a-2b, 1a-2c, 1b-2b, 1b-2c, 1c2b,

1c-2c, 1d-2b, 1d-2c, 1e-2b,

1e-2c, 1f-2b, 1f-2c, 1g-2b, 1g-2c

State genomic activities

Knowledge and interest

Continuous

Categorical

1h-2a

State genomic activities

Knowledge and interest

Continuous

1h-2b, 1h-2c

Data Analysis

All survey data were obtained through the Qualtrics survey platform and directly

exported into IBM SPSS version 23.0 software to perform statistical analysis, which is

the most appropriate platform for analysis of quantitative survey data (IBM, 2016). The

raw data was only available to the committee and myself and first examined for and

cleaned of incomplete or duplicate entries or any other potential abnormalities. Assigned

variable names and numerical values were also transferred in the statistical report

downloaded for data analysis. Numerical values for “Don’t know” responses were

changed to “2” to assure they were not included in the statistical analysis and treated as

“No” (not present) initially for this study. “Don’t know” values were then returned to

their original numbers to see what impact, if any, they had on the analysis.

Descriptive statistics were performed first to identify means, standard deviations,

and range of values for all variables (Creswell, 2009). Construction of frequency

distributions looked at response patterns for all variables with nominal questions showing

modes and interval questions providing median, mean, range, standard deviation, and

coefficient of variation (Frankfort-Nachmias & Nachmias, 2008). Frequencies were

converted into percentages for meaningful interpretation and comparison and visually

displayed in the results section through tables and graphs.

Research Question 1

To what extent, if any, is there an association between states that have received

funding for chronic disease genomics and the level of knowledge and interests in

genomics by state and territorial CDDs?

H01: There is no association between states that have received funding for chronic

disease genomics and the level of knowledge and interests in genomics by state and

territorial CDDs.

H11: There is an association between states that have received funding for chronic

disease genomics and the level of knowledge and interests in genomics by state and

territorial CDDs.

To determine possible associations of the variables in Research Question 1, I used

Chi-square analysis, the independent t test, and multiple linear regression. Chi-square

analysis examined possible associations between current state genomic funding and

whether CDDs knew how to contact genetics professionals if they needed to refer a

patient or required professional consultation. The independent t test was used to analyze

whether or not state genomic funding had a possible association with CDDs level of

knowledge and interest in genomics determined on a continuous scale. Finally, multiple

linear regression determined possible associations using current state genomic funding as

the independent variable and the level of knowledge and interest in genomics by CDDs

(determined on a continuous scale) as the dependent variable.

Research Question 2

To what extent, if any, is there an association between current state genomic

activities and the level of knowledge and interests in genomics by state and territorial

CDDs?