Defining Prevention Science

Zili Sloboda and Hanno Petras (eds.)Advances in Prevention ScienceDefining Prevention Science201410.1007/978-1-4899-7424-2_1

© Springer Science+Business Media New York 2014

1. Prevention Science: An Epidemiological Approach

David Cordova¹  , Yannine Estrada², Shandey Malcolm², Shi Huang², C. Hendricks Brown², Hilda Pantin² and Guillermo Prado²

(1)

School of Social Work, University of Michigan, 1080 S. University, Suite 2846, Ann Arbor, MI 48109, USA

(2)

Department of Epidemiology and Public Health, Miller School of Medicine, University of Miami, Miami, FL, USA

David Cordova

Email: [email protected]

Abstract

This chapter provides an overview of prevention science and the role of epidemiology in the field of prevention science. Specifically, we discuss several major ways in which prevention science is informed by epidemiology. First, we describe how epidemiology is useful in identifying target populations and vulnerable periods as well as discuss the distribution of disease, etiological risk and protective factors, and human development across the lifespan. Second, we highlight the ways in which epidemiology is used to develop frameworks, including an ecological and developmental framework to prevention, and how these frameworks are useful for understanding and comparing targeted populations for specific times of risk. Third, we describe experimental epidemiology and widely used analytic and methodological approaches for testing the efficacy and effectiveness of interventions. Fourth, we describe the role of epidemiology in implementation strategies. Finally, we discuss the need to work toward feedback loops whereby prevention science findings are used to inform epidemiology and vice versa. Although the fields of prevention science and epidemiology share a common goal and interest in health promotion and disease prevention, to some extent, both disciplines often operate in isolation with a minimal feedback loop process. These feedback loops may be essential in advancing both fields. To highlight these concepts, we use several disease and health-risk behaviors, including HIV/AIDS, obesity, and alcohol and drug use.

Introduction

This chapter provides an overview of prevention science and the role of epidemiology in the field of prevention science. Specifically, we discuss several major ways in which prevention science is informed by epidemiology. First, we describe how epidemiology is useful in identifying target populations and vulnerable periods as well as discuss the distribution of disease, etiological risk and protective factors, and human development across the lifespan. Second, we highlight the ways in which epidemiology is used to develop frameworks, including an ecological and developmental framework to prevention, and how these frameworks are useful for understanding and comparing targeted populations for specific times of risk. Third, we describe experimental epidemiology and widely used analytic and methodological approaches for testing the efficacy and effectiveness of interventions. Fourth, we describe the role of epidemiology in implementation strategies. Finally, we discuss the need to work toward feedback loops whereby prevention science findings are used to inform epidemiology and vice versa. Although the fields of prevention science and epidemiology share a common goal and interest in health promotion and disease prevention, to some extent, both disciplines often operate in isolation with a minimal feedback loop process. These feedback loops may be essential in advancing both fields. To highlight these concepts, we use several disease and health-risk behaviors, including HIV/AIDS, obesity, and alcohol and drug use.

Introduction to Prevention Science and Epidemiology

Prevention science is concerned with, among other things, identifying antecedents that impact health and health behavior, as well as the development of models to ameliorate undesired health outcomes and promote health behavior. A relatively new field, prevention science is interdisciplinary and combines life-course development, community epidemiology, and preventive intervention perspectives. Epidemiology in particular plays an important role in prevention science. Below, we describe the role of epidemiology in the field of prevention science.

Traditionally, epidemiology has been defined as the study of the distribution and determinants of health; its aim is to prevent and reduce, survey, and control health disorders (Susser & Stein, 2009). As the cornerstone method of public health research, epidemiology plays a key role in prevention science. Of particular importance to the field of epidemiology is describing the natural history or career of the outcome of interest. The natural history can be defined as the progression of the outcome of interest from the time of exposure to cessation/desistance (Gordis, 2009). From this perspective, central to epidemiology is to identify those factors that contribute to the onset, progression/escalation, and cessation/desistance of the behavior or problem of interest. Using substance use as an example, epidemiology is interested in identifying factors contributing to the onset of substance use, risk and protective factors, whom to target and at what developmental stage, and the methods and design that may be most optimal for highlighting the natural history of substance use. For example, epidemiologic studies show us that early onset of substance use among adolescents, an important period of development, is associated with a greater likelihood of later abuse and dependence (Behrendt, Wittchen, Höfler, Lieb, & Beesdo, 2009; Grant & Dawson, 1998), as well as identify those correlates and risk factors that may contribute to the escalation or progression from use to abuse and dependence (Swendsen et al., 2012).

Epidemiological concepts, including the distribution of disease, determinants of health, and the role of host-environment-agent in understanding disease and etiology, are important in advancing the field of prevention science. These concepts are described below.

The distribution of disease refers to the frequency, pattern, and history of a particular condition, disorder, or disease among groups or populations. Identifying the distribution of disease is helpful in determining segments of the population that are most affected by a disease, condition, or disorder and plays an important role in informing the development of preventive interventions for specific subpopulations, including racial/ethnic minorities, developmental age groups, and/or those who live in certain geographic regions. For example, epidemiologic research indicates that, when compared with their non-Hispanic White and African American counterparts, Hispanic youth report the highest lifetime, annual, and 30-day prevalence rates of both licit and illicit substance use, excluding amphetamines (Johnston, O’Malley, Bachman, & Schulenberg, 2013). These epidemiologic research findings indicate that specific populations may require specialized prevention services, including the development of Hispanic-specific preventive interventions (Prado et al., 2007, 2012). The distribution of disease, however, does not operate in isolation and is influenced by genetic, environmental, and social determinants of health that also involve risk and protective factors (Susser & Stein, 2009).

The determinants of health can be defined as those risk and protective factors that may have an impact on the distribution of disease and occurrence of a condition or an event (Torrence, 1997). Models of causation, including the host-environment-agent, are useful for better understanding the role of risk and protective factors associated with physical and psychological health disorders.

The host-environment-agent model is used to describe the intersection of the host (e.g., an individual), the environment (e.g., vehicle), and agent (e.g., driver distraction) that interact with one another in the development of a condition such as motor vehicle injuries. The host is the individual and his/her inherent characteristics (e.g., genetic, psychological), which may be a precursor for the development of a particular health or psychological condition. Across different prevention fields, the role of agent can vary and refers to the organism or direct cause of the condition. Using motor vehicle injuries and driver distraction as an example, the agent could include the use of mobile communication devices. The environment includes all external factors that may contribute to the development of a condition and interacts with an individual’s susceptibility. For example, accessibility of mobile communication devices, cultural norms and practices, and legislation on the use of mobile communication devices while driving are environmental factors that may impact a driver’s use of mobile communication devices and consequently motor vehicle injuries and through which risk and protective factors can be targeted by preventive interventions. In fact, prevention science findings have highlighted the significant consequences and public health concern of using mobile communication devices while driving and motor vehicle injuries (Ibrahim, Anderson, Burris, & Wagenaar, 2011). For example, findings from the 2008 Fatality Analysis Reporting System (National Highway Traffic Safety Administration, 2009) indicate that approximately 16 % of all reported fatal crashes are attributable to driver distraction, including the use of mobile communication devices. Although it is still unclear how restricting the use of mobile communication devices while driving impacts motor vehicle injuries, these data have led to the implementation of preventive legislation (i.e., environment) on the use of communication mobile devices (i.e., agent) while driving (i.e., individual, host) in 39 states and the District of Colombia (Ibrahim et al., 2011). Thus, the host-environment-agent model has helped guide the field of prevention science and the development and evaluation of motor vehicle injury legislation prevention models to target risk and protective factors within the host, environment, and agent domains (Ibrahim et al., 2011). Therefore, etiological frameworks, including the host-environment-agent model, facilitate advances in the field of prevention science by providing a conceptual framework for better understanding problems, diseases, and disorders; selecting settings and stages of life that may be particularly conducive to intervention; and guiding the choice of what constructs to measure.

To summarize, epidemiology aims to contribute to the prevention of health disorders and health-risk behaviors by providing models through which the natural history of disease, along with the potential risk and protective factors associated with it, can be described and defined (Torrence, 1997). As shown in Fig. 1.1, the roles of the distribution of disease, determinants of health, and the host-environment-agent are key first steps in understanding disease and etiology, as well as informing the development of prevention programs. In the next section, we describe the role of etiological models in the development of preventive interventions.

Fig. 1.1

The role of epidemiology in prevention science. Adapted from Sloboda, Glantz, and Tarter (2012)

Etiology

A relatively new perspective has emerged shifting focus to the developmental vulnerability of individuals in context. This paradigm shift allows us to better understand the etiology of the outcome of interest, as well as who will and who will not engage in certain risky and unhealthy behaviors (Sloboda et al., 2012). Specifically, the etiology of health-risk behaviors, such as adolescent drug abuse, may best be understood through a risk and protective factor model, informed by a developmental framework, that examines the intersection of genetic, psychological, and environmental factors (Prado et al., 2009; Sloboda et al., 2012). As early as the mid-1960s, researchers began to identify a host of risk and protective factors associated with behavioral health problems (Berrueta-Clement, Schweinhart, Barnett, & Weikart, 1987). Identified risk and protective factors include individual-level (e.g., genetic, psychological), proximal environmental (e.g., family, school, peers, and work), and distal environmental factors such as community and social/economic conditions (Kellam & Langevin, 2003). Furthermore, researchers state that risk and protective factors do not operate in isolation and should be studied as an integrated developmental process that also includes genetic factors (Schwartz, Pantin, Coatsworth, & Szapocznik, 2007). Risk and protective factor models have great utility to the field of prevention science, particularly as it relates to identifying different points of intervention that prevention programs may target.

Genetics

Research has highlighted the ways in which genetic markers play an instrumental role in the predisposition to engage in, for example, alcohol and drug use. In fact, studies examining heritability estimates suggest that genetics account for as much as 40–70 % of the risk and susceptibility in the development of alcohol or drug abuse and dependence (Heath et al., 1997). Thus, prevention science may play an important role in ameliorating health-risk behaviors, including alcohol or drug abuse and dependence, particularly through a better understanding of the process of genetic markers. For example, it is hypothesized that adolescents with one or two copies of the short variant 5HTTLPR genetic marker may be more susceptible to engaging in health-risk behaviors, relative to adolescents with two copies of the long variant 5HTTLPR genetic marker. Prevention scientists have demonstrated that participating in a preventive intervention may mitigate the risk of the 5HTTLPR genetic marker, thereby ameliorating the genetic predisposition for engaging in health-risk behaviors (Brody, Beach, Philibert, Chen, & Murry, 2009; Brody, Chen, Beach, Philibert, & Kogan, 2009). Therefore, disease and problem behaviors are best understood in the context of the interplay between genetics and environment. For example, gene × environment approaches have played a significant role in advancing the field of prevention science, including working toward a better understanding of genetic susceptibility and cigarette smoking in people with Crohn’s disease (Helbig et al., 2012) and gene × lifestyle and gene × drug interactions and obesity (Franks & Poveda, 2011). Advances in technology, methodology, and statistics will further our understanding of the role of gene × environment in prevention science, as well as advance rodent models to human models to identify the disease or problem behavior’s susceptibility pathways (Blackburn & Jerry, 2011). The gene × environment model, therefore, is very promising with respect to advancing the etiology literature, which in turn has the potential to maximize the efficacy and effectiveness of prevention programs (Howe, Beach, & Brody, 2010).

Psychological Factors

In addition to genetic factors, psychological or intrapersonal factors are important in understanding health-risk behaviors. Psychological or intrapersonal factors refer to those cognitive, emotional, and attitudinal processes that vary across individuals and can serve as risk or protection for health and behavior (Hemphill et al., 2011; Tonin, Burrow-Sanchez, Harrison, & Kircher, 2008). For example, adolescents who report positive attitudes toward alcohol or drug use may be more vulnerable or susceptible to engaging in alcohol or drug use compared with youth who report more negative attitudes toward alcohol or drug use (Cordova et al., 2011; Hemphill et al., 2011; Tonin et al., 2008). To this end, prevention science may play an important role in ameliorating negative psychological or intrapersonal factors. For example, cognitive and attitudinal processes are pathways through which prevention programs have demonstrated efficacy in preventing/reducing alcohol and drug use among adolescents (Hops et al., 2011).

Environmental/Ecological Factors

Environmental/ecological factors also play a role in the development of health conditions. Environmental/ecological factors are those social, cultural, and contextual processes, including family, peer, community, and legislation, that both influence and are influenced by the individual (Bronfenbrenner, 1979, 1989; Catalano & Hawkins, 1996; Szapocznik & Coatsworth, 1999). A substantial amount of research has demonstrated the ways in which environmental/ecological factors influence risk or protection for health and mental health (Catalano & Hawkins, 1996; Cleveland, Feinberg, & Greenberg, 2010; Griffin & Botvin, 2010). For example, studies suggest that communities characterized by higher levels of social capital and neighborhood collective efficacy may have protective effects on depression among certain Hispanic subgroups (Vega, Ang, Rodriguez, & Finch, 2011). Thus, the environmental/ecological context is important in better understanding risk and protection for health and mental health. Prevention programs that target environmental/ecological processes at multiple contextual levels may be particularly helpful in decreasing health-risk behaviors and promoting health (Hawkins et al., 2012). In fact, Hawkins and colleagues (2012) have shown that, when compared with youth in control communities, youth in Communities That Care, a community-based prevention program, demonstrate a decrease in community-wide levels of health-risk behaviors, including substance use over time.

Etiology Across the Lifespan

The role of genetics, psychological, and environmental/ecological factors on health and health-risk behaviors should be informed through a developmental and life-course lens (Kellam et al., 1991; Sloboda et al., 2012). For example, we know that individuals might be at increased risk for alcohol and drug use in certain developmental periods across the human development life cycle. In fact, research has shown that adolescence and early adulthood in particular are developmental stages in which drug abuse might be more pronounced and likely to occur, compared with middle adulthood (Botvin, Griffin, Paul, & Macaulay, 2003; Dishion, Kavanagh, Schneiger, Nelson, & Kaufman, 2002; Hawkins, Catalano, & Miller, 1992). Therefore, adopting a developmental and life-course perspective to inform prevention science might be important in interrupting the sequelae of negative health and psychological outcomes during critical periods across the lifespan (Braveman & Barclay, 2009).

Developmental Epidemiology

Informed by both a developmental and life-course perspective, developmental epidemiology identifies specific proximal individual or environmental factors at an early stage of life to then target preventive interventions at these factors (Kellam & Langevin, 2003). Interventions are then evaluated to determine whether and the extent to which targeting the identified risk factors has positively impacted more distal factors across both time and development. Critical periods are an important concept in developmental epidemiology and are essential to describing the individual’s life course (Braveman & Barclay, 2009). A critical period refers to a window of time during the life course when a given exposure has a critical or even permanent influence on later health (Braveman & Barclay, 2009, p. S164). Therefore, a critical period refers to those critical benchmarks, including age, gender, and cultural relevance and expectations that might interact during certain times of transitions across the life course (Chambers, Taylor, & Potenze, 2003; Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003). Here, we use the Good Behavior Game (Kellam et al., 2008) prevention program as an example. A developmentally informed and universal classroom behavior management prevention model, the Good Behavior Game (Kellam et al., 2008) targets first and second grade children and has demonstrated long-term positive effects, particularly during critical periods (e.g., late adolescents and emerging adulthood), on several health outcome indicators, including substance use and delinquency. The choice of its use on entry into elementary school was based on prospective, longitudinal studies that identified aggressive/disruptive behavior in first grade as a strong antecedent for adolescent drug and alcohol use and abuse/dependence in young adulthood (Kellam et al., 2008). Thus, the development of prevention programs such as the Good Behavior Game (Kellam et al., 2008), which aim to combat problem behaviors early and during a developmental stage that has been identified as a critical period for subsequent drug use, is an example of how developmental epidemiology may be used in informing prevention services.

Life Course/Social Field Theory

Building on developmental epidemiology, life course/social field theory posits that individuals are embedded in social fields or contexts that require social task demands. These social task demands have criteria for success or failure and vary based on developmental stages of life as well as critical transition periods across each developmental stage of life (Kellam & Van Horm, 1997). For example, learning how to drive responsibly is one social task demand that many experience during late adolescence and early adulthood in the United States. The life course/social field framework provides prevention scientists with a platform to integrate various disciplines (Kellam & Van Horm, 1997).

Genetic, psychological, environmental/ecological, developmental, and life-course factors provide a risk and protective framework for understanding complex human behaviors and the ways in which these factors both influence and are influenced by one another over time (Kaufman et al., 2007; Sloboda et al., 2012). Furthermore, descriptive epidemiology constitutes a key first step to understanding the disease or problem behavior and provides prevention scientists with the necessary tools to identify both the frequency and the distribution of risk and protective factors in populations as well as to assess the extent of a disease or problem behavior (Gordis, 2009). Analytical epidemiology then uses this information to examine, for example, these risk and protective factors to better understand the etiology of the outcome of interest. Next, we describe the role of etiological frameworks in guiding the development of preventive interventions.

The Role of Etiological Models in Prevention Science

Several etiological models have been developed to highlight the role of genetic, psychological, environmental, and developmental factors on risk behaviors, including Bronfenbrenner’s (1979, 1989) ecological systems theory. From this perspective, risk behaviors are a function of both the individual and the environment. Ecological systems theory (Bronfenbrenner, 1979, 1989) conceptualizes development as taking place within contexts, namely, the microsystem, mesosystem, exosystem, and macrosystem. The microsystem consists of risk and protective factors proximal to the individual. For example, risk factors for childhood obesity in the microsystem include the type of nutrition provided in the home. The mesosystem refers to the relationship and processes between two microsystems. Factors in the mesosystem, for example, may include parental involvement that supports healthier meal options or choices in a child’s school context. The exosystem are those systems in which an individual does not directly participate but that can have an impact on the individual. For example, parental work conditions (e.g., long work hours) might prevent parents from providing a healthy dinner at home, which in turn may lead to children having to prepare processed foods and thereby have an effect on the child’s eating habits. Lastly, the macrosystem encompasses those variables present in the broader social and cultural systems such as a policy requiring more transparent nutritional food labeling for consumers. Etiological models, including the ecological systems theory (Bronfenbrenner), have helped advance the field of prevention science and develop next generation etiological models.

Building on Bronfenbrenner’s work, several next generation ecological frameworks have been developed to conceptualize the role of environmental/ecological factors on the etiology of risk and protection and advance the field of prevention science, including the ecodevelopmental theory (Szapocznik & Coatsworth, 1999). Ecodevelopmental theory (Szapocznik & Coatsworth, 1999), for example, is helpful in conceptualizing integrated developmental risk and protective processes operating in the lives of individuals (Pantin et al., 2003; Prado et al., 2010; Szapocznik & Coatsworth, 1999). Ecodevelopmental theory (Szapocznik & Coatsworth, 1999) affirms that social domains in which the individual is embedded both influence and are influenced by the individual in a developmental context and occurs on various levels, including those microsystems, mesosystems, exosystems, and macrosystems described above. From this viewpoint, health behaviors are influenced by a multiplicity of factors, some of which are proximal to the individual, whereas others are distal.

The development of theoretical models, such as ecodevelopmental theory (Pantin et al., 2003; Prado et al., 2010; Szapocznik & Coatsworth, 1999), serve a vital role to the field of prevention science and epidemiology. Theoretical models help prevention scientists understand etiology, aid in predicting health-risk behaviors among the populations we work with, and help ascertain the pathways through which prevention programs work (Brown et al., 2008, 2009). Conversely, the fields of prevention science and epidemiology also play an important role in developing and testing theoretical models, at both the population level and the more targeted high-risk groups.

Prevention science and epidemiology inform what population-based strategies (i.e., universal), as compared with high-risk targeted strategies (i.e., indicated), are theoretically able to accomplish (Brown & Faraone, 2004). We take, for example, the prevention of drug abuse in adolescents. LifeSkills Training, a universal school-based prevention program found to be effective in preventing/reducing drug use, was designed for all ethnic and racial students within a particular school setting (Griffin, Botvin, Nichols, & Doyle, 2003). This drug use prevention strategy has the potential to reach a large segment of students, and therefore it is a population-based or universal strategy (Brown & Faraone, 2004). In comparison, identifying drug abuse risk factors can also lead to the development of interventions that target a specific segment of the population that shares a particular risk factor (Brown & Faraone, 2004). For example, Familias Unidas (Pantin et al., 2009; Prado et al., 2007; Prado & Pantin, 2011), a Hispanic-specific, family-based preventive intervention that targets identified risk factors that may be more pronounced in Hispanic families (e.g., cultural differences and parent–adolescent communication), has been found to be efficacious in preventing/reducing drug use among Hispanic adolescents. Thus, prevention science can be seen as an epidemiologic experiment because it aims to determine whether and the extent to which prevention programs and population-based or high-risk strategies target identified etiological and theoretical predictors, as well as identify which processes account for the differences caused by the intervention.

As shown in Fig. 1.1, epidemiology and theoretical models inform the field of prevention science and the development of preventive interventions. Prevention programs in turn can be used to test these models and help advance and adapt theoretical frameworks as a result of new evidence and knowledge gained. In the next section, we describe some of the epidemiologic methods and study designs used to evaluate prevention programs.

The Use of Epidemiologic Methods and Study Design in Evaluating Preventive Interventions

Advanced longitudinal epidemiologic methods and study designs have provided prevention scientists with the necessary tools to efficiently and more effectively evaluate preventive interventions and determine to what extent the effects are sustained over time (Hayes, 2006). These tools facilitate moving a program of research from efficacy to scale and promote the use of prevention research findings in the advancement of epidemiologic models (Kellam et al., 2011). Longitudinal methods and research designs examine change over time by following individuals beyond the period when they are actively participating in prevention or early intervention efforts and thereby provide valuable insight into the sustainability of prevention programs. Change and trajectory of outcomes over time can now be conducted because of advancements in epidemiologic data collection methods and advanced statistical methods that allow for the testing of more complex etiological and theoretical models (McArdle, 2009). Thus, advanced longitudinal methodologies inform the effects of preventive interventions, both short and long term, promote the advancement of epidemiological and prevention models, and work toward best practices. Before a prevention program has been found to be an evidence-based or best practice model of prevention, however, the program must first undergo rigorous scientific testing and go through a research process that includes basic science, efficacy, effectiveness, implementation, and taking interventions to scale (Brown et al., 2008, 2009; Van Spall, Toren, Kiss, & Fowler, 2007). Basic science and epidemiology are used to develop etiological and theoretical models, which in turn can be used to examine the efficacy of a prevention program. The efficacy of a prevention program is determined by whether and the extent to which it works under ideal conditions. If and when a program is found to be efficacious, then effectiveness trials, which examine the effects of prevention programs in real-world settings, can take place. Once established as an effective model, prevention programs can work toward implementation and scale to ensure widespread adoption. It should not be surprising that epidemiological and prevention science methods are integrated throughout this process. For example, theoretical models may be adapted, or prevention programs tailored, based on the results of outcome and process data, which help optimize efficacious or effective models. Additionally, evaluation is particularly important throughout the research process. Several epidemiological research methods are used to evaluate the effects of prevention programs, including time to event, growth curve, multilevel modeling, mixture modeling, and mediation modeling. Below, we discuss some of these widely used methods in evaluating prevention interventions.

Of importance to the field of prevention science is the evaluation of prevention programs, and prevention trials methodology has developed as a particularly important type of experimental epidemiology (Brown et al., 2008). This is accomplished through the testing of etiological and intervention models that help guide prevention services, as well as identifying those pathways through which preventive interventions work. This is important not only to gain a better understanding of intervention processes but also to confirm that the intervention is in fact targeting the hypothesized etiological factors and is casually related to the outcomes of interest. Therefore, prevention science is concerned with, among many other things, answering the question: How do preventive interventions work, for whom, and under what cultural, social, and institutional conditions (Brown et al., 2008; Kellam et al., 2011; Tein, Sandler, MacKinnon, & Wolchik, 2004)?

Given that the primary goal of epidemiology is to describe the natural history of the outcome of interest, advanced statistical methods, including time-to-event, growth-curve, and multilevel models, are especially important for measurement of change over time and thereby more accurately describe the natural history process. For example, time-to-event analysis provides prevention scientists with the tools to capture, to some extent, the natural history of an outcome of interest, including the onset of the behavior of interest, progression/escalation, and the cessation/desistance of that behavior. Multilevel modeling, for example, is used to describe how individuals change over time as well as the extent to which the changes may vary across individuals (Singer & Willett, 2003). In growth-curve modeling, we can examine individual trajectories, individual differences in these trajectories, predictors of individual differences, and their effects on development over time. Furthermore, growth-curve modeling can describe important group statistics to better understand developmental processes at the group level (Duncan, Duncan, & Strycher, 2006). For example, prevention scientists interested in examining whether and the extent to which a prevention program is efficacious in preventing an outcome of interest, relative to a control condition, may find the growth-curve modeling approach useful. Here, we can describe trajectories of individuals both within and across conditions. Additionally, mixture modeling, including latent class growth analysis (LCGA) and growth mixture modeling (GMM), allows prevention scientists to capture classes or subpopulations that are unknown (Muthen & Muthen, 1998–2010). For example, mixture modeling may be helpful in identifying those classes or subpopulations that benefit most from participating in a drug abuse prevention program.

Some research has been conducted to highlight for whom preventive interventions work and for whom they do not work. For example, Hispanic adolescent preventive interventions may be more efficacious among US-born youth, relative to foreign-born youth, on some behavioral and mental health outcomes (i.e., moderator; Cordova et al., 2011; Martinez & Eddy, 2005). Additionally, research has demonstrated the ways in which the effects of preventive interventions vary by gender (i.e., moderator; Kulis, Marsiglia, Ayers, Calderón-Tena, & Nuño-Gutiérrez, 2011; Kulis, Yabiku, Marsiglia, Nieri, & Crossman, 2007). Furthermore, research has been conducted to identify how prevention programs work. For example, we now know that family functioning is one pathway through which family-based, drug abuse preventive interventions are efficacious (i.e., mediator; Pantin et al., 2009; Prado et al., 2007; Sandler, Schoenfelder, Wolchik, & MacKinnon, 2011). Although some research has focused on answering for whom and how preventive interventions work, little research has been done to identify research processes associated with the cultural, social, and institutional conditions that may have an effect on prevention programs (Kellam et al., 2011; Sandler et al., 2011). Working toward a better understanding of these research processes might have great utility in identifying potential factors that contribute to prevention intervention outcomes. This knowledge may elucidate optimally efficacious and effective preventive interventions, which in turn could aid in adapting or extending etiological and theoretical models (Kellam et al., 2011). One widely used model for better understanding research processes and testing theoretical frameworks in the field of prevention science is the mediation model.

The mediation model examines whether and the extent to which a third variable has an effect on the relation between two other variables (Brug, Oenema, & Ferreira, 2005). Theoretical frameworks are especially important in the selection of mediating processes and predictors because they, by definition, are informed by a large body of scientific research and can provide strong evidence for the etiology (e.g., risk and protective factors) of a disease. In addition to facilitating the selection of mediation processes, theoretical frameworks present a systematic way of understanding causal determinants of disease and how these factors may operate to reduce disease, which may be helpful in advancing the fields of epidemiology and prevention science. Once a viable mediation process has been identified, a hypothesis can then be generated that will guide the research protocol, including the pathways (e.g., family functioning) to target and how change processes occur. Thus, a mediation analysis can test a theory in intervention; to the extent that changes in a hypothesized mediator predict distal outcomes, the theoretical model is confirmed (MacKinnon & Luecken, 2008).

In addition to the importance of statistical methods in evaluating preventive interventions, equally important is the study design. Below, we describe several study designs that are helpful in evaluating prevention programs.

The randomized controlled trial (RCT) can be particularly helpful in moving forward the prevention science field because RCTs, in theory, generate equivalent condition groups that in turn allow prevention scientists to say with greater certainty that prevention effects are attributable to the intervention itself (Jadad et al., 1996; Olsen, Christensen, Murray, & Ekbom, 2010; Silverman, 2009). Although RCTs are the optimal strategy in the field of prevention science, they may not be feasible in the evaluation of all preventive interventions (Silverman, 2009). For example, there may be circumstances in which participants cannot plausibly and ethically be randomized to an experimental or control condition; consider faith-based interventions among persons who may not practice organized religion (Rubin, 1974, 2005). Therefore, several preventive intervention evaluation alternatives exist that may be more appropriate for examining the effects of prevention programs, including observational, rolled-out, quasi-experimental, and randomized encouragement designs.

Observational methods may be particularly helpful for better understanding disease in instances where randomization of participants is impossible or unethical or in the outbreak of a rare disease. Observational methods, including the cohort design, allow prevention scientists to follow a segment of the population that may not have the disease or disorder of interest and observe this sample as a natural ecological experiment. That is, through longitudinal and correlation analyses and life histories, the cohort design allows prevention scientists to examine risk and protective factors associated with that particular disease or disorder in a natural setting (Hemingway & Marmot, 1999; Stroup et al., 2000). A limitation to observational studies, however, is that they introduce several types of biases, including selection bias when the choice of control groups may differ systematically from intervention groups (Benson & Hartz, 2000; Hemingway & Marmot, 1999).

Another alternative to the traditional RCT design is the rolled-out or dynamic wait list design (Brown, Wyman, Guo, & Peña, 2006; Brown et al., 2009). In this design, an intervention that has already been selected for wide-scale implementation is evaluated. For instance, a government mandate may require that all schools within a specific district deliver a sex education program. The sex education program may have limited empirical support, but an opportunity exists to evaluate the program as it is rolled-out into the school district. In the rolled-out methodology, all of the study units (individuals or clusters) receive the intervention but are randomly assigned to a time interval in which to receive the intervention. More specifically, all of the study units begin in the control condition and, as participants receive the preventive intervention services, participants are switched to the experimental condition. This process occurs until all of the study units eventually receive the intervention. Finally, the effects of the experimental condition on the outcomes are examined and compared with the control condition. This process occurs until the last interval, when all study units have received the intervention and comparisons can no longer be made. There are several benefits to this type of design. For instance, the rolled-out method is advantageous in circumstances where ethical concerns or public health mandates do not support the exposure of an intervention to only a subset of individuals. The rolled-out method is also advantageous for practical reasons, for instance, when financial resources limit the ability to deliver the intervention simultaneously to all study participants (Brown et al., 2006, 2009).

Nonrandom or quasi-experimental trials are also alternatives to RCTs. Several types of quasi-experimental designs exist. For example, in time-series quasi-designs, a single group of study participants is assessed both before and after the prevention intervention activities. A significant favorable outcome between pre- and post-intervention, to the extent that it is not due to chance, suggests an efficacious/effective intervention (Grimshaw, Campbell, Eccles, & Steen, 2000).

Randomized encouragements are epidemiologic designs in which participants are randomized not to a preventive intervention but rather to an opportunity or encouragement to take part in an intervention (Vinokur, Price, & Schul, 1995). In a smoking cessation prevention program, for example, participants may be randomly assigned either to receive an invitation or not to receive an invitation. In this methodology, it is assumed that the impact of the assignment is mediated entirely through the receipt of the treatment. This process aims to remove the necessity to fully adhere to the prevention protocol, as is the case in RCTs. Finally, in nonrandom quantitative assignment designs, participants are not randomized but rather assigned to treatment based on a quantitative measure such as financial need (Finkelstein, Levin, & Robbins, 1996a, 1996b).

Additionally, epidemiological study designs include case control studies, cohort-age-period designs, twin studies, and cross designs. Case control studies, for example, have great utility in situations where the disease of interest is rare, observing a sample or control series of the population. This sample is then used in place of the complete assessment of disease frequencies, including the proportion, rate, and odds (Rothman, Greenland, & Rash, 2008).

Applying Prevention Research Findings to Epidemiology: Informing the Advances of Epidemiologic Models

We have described the ways in which epidemiological data have great utility guiding and informing the development and evaluation of preventive interventions. For example, epidemiological data show that HIV risk behaviors are a complex phenomena occurring at multiple levels (DiClemente et al., 2004; Prado, Pantin, Schwartz, Lupei, & Szapocznik, 2006), which in turn has called for the development and evaluation of HIV preventive interventions at multiple levels, including family (Villarruel, Jemmott, Jemmott, & Eakin, 2006), peer (Guarino, Deren, Mino, Sung-Yeon, & Shedlin, 2010; Raja, McKirnan, & Glick, 2007), school (Nkansah-Amankra, Diedhiou, Agbanu, Harrod, & Dhawan, 2011), and community (Harper, Bangi, Sanchez, Doll, & Pedraza, 2009) to help curb the tide of HIV infection. Although the use of epidemiologic findings to inform the development and evaluation of preventive interventions has been widely practiced, little has been done with respect to using prevention findings to inform epidemiology (Sloboda et al., 2012). That is, the use of scientific knowledge produced from the implementation and evaluation of prevention programs to inform epidemiology has been minimal. Working toward epidemiologic models that are informed by prevention findings might prove useful in advancing the fields of epidemiology and prevention science, particularly with respect to public health. Therefore, the contribution of prevention science for epidemiology is then to better specify when to measure, what to measure, whom to measure, what to target, and more specifically to assess the viability of the dominant perspective on the etiology of the natural history of the outcome.

To highlight one way that prevention findings can inform epidemiology, we refer back to the mediation model and use the Familias Unidas (Pantin et al., 2009; Prado et al., 2007; Prado & Pantin, 2011) preventive program as an example. As previously mentioned, family functioning is the pathway through which Familias Unidas has an effect on both drug use and unprotected sex in multiple efficacy studies (Pantin et al., 2009; Prado et al., 2007; Prado & Pantin, 2011). The question still remains, however: Will family functioning remain the significant pathway through which Familias Unidas operates if/when the program moves to implementation? For example, it may very well be that other factors such as community processes, including community partnerships and networks, may be the pathway through which the Familias Unidas program has an effect on the outcomes. This in turn can be used to inform epidemiologic models. To that end, Kellam and colleagues (2011) have proposed several strategies that may be helpful to the process of prevention findings informing epidemiologic models and the field of prevention science, including community partnerships and networks (Kellam et al., 2011).

Community partnerships are essential to the research process and for the development of the next generation of prevention research. For example, preventive interventions are effective only to the extent that they are accepted and implemented with fidelity by the community. Thus, building and sustaining community partnerships for prevention efforts are strategies through which prevention science findings may inform epidemiologic models, which in turn may advance the field of prevention science (Kellam et al., 2011). For example, community partnerships can inform prevention scientists whether and the extent to which the prevention program’s aims and goals are shared with that of the communities’ aims and goals (Kellam et al., 2011). This information can then be used to tailor prevention services and epidemiological models and work toward optimally efficacious and effective programs.

Prevention research findings may also help advance epidemiologic models, particularly when moving a prevention program from efficacy to effectiveness to implementation (Kellam et al., 2011). To do so, networks should be established that consist of researchers, policy makers, and practitioners with a shared goal of working toward translational research (i.e., efficacy to effectiveness to implementation). To this end, networks can foster the sharing of diverse research experiences, including identified theoretical models, measures, and relevant assessment for a particular population and context (Kellam et al., 2011).

A prevention model that has been shown to be effective, particularly as it relates to implementation strategies, is the RE-AIM framework (Glasgow, McKay, Piette, & Reynolds, 2001; Glasgow, Vogt, & Boles, 1999). The RE-AIM framework (Glasgow et al., 1999, 2001) is a translational research model—moving efficacious programs to real-world settings—and is concerned with factors and processes related to long-term effectiveness impact. The RE-AIM framework is guided by five steps: reach, efficacy, adoption, implementation, and maintenance (Glasgow et al., 1999, 2001). Reach refers to engagement, active participation, and retention in prevention services, which are affected by contextual barriers to participation, including transportation, work schedule, and child care. Efficacy is concerned with the effects of a prevention program on the outcome variables when it is implemented with fidelity and in a controlled setting. Adoption can be described as the process by which a best practice program is delivered at the system level and the proportion of organizations that are willing to adopt a particular best practice model. Implementation operates at the real-world level and is concerned with the consistency of delivery and translation of a best practice model. Finally, maintenance is concerned with the long-term effects of prevention programs on behavior and how the best practice is institutionalized and put into practice (Glasgow et al., 1999, 2001). The Nurse Family Partnership (Olds, 2008; Olds, Henderson, & Kitzman, 1994; Olds, Henderson, Kitzman, & Cole, 1995) prevention program is an example of a translational program of research that has implemented the RE-AIM framework and has been rigorously tested over the past 20 years (Donelan-McCall, Eckenrode, & Olds, 2009). Nurse Family Partnership (Olds, 2008; Olds et al., 1994, 1995), a home-visiting prevention intervention, is efficacious in preventing/reducing child maltreatment, including child abuse and neglect, emergency room visits, and mothers’ substance use during gestation. First evaluated in 1977 in an RCT, Nurse Family Partnership has spanned a program of research from reach to efficacy to adoption to implementation to maintenance (Donelan-McCall et al., 2009). In fact, the Nurse Family Partnership now houses a national service office that assists communities with replication and ensuring fidelity in the implementation of the Nurse Family Partnership program. Here, the Nurse Family Partnership provides a model to the field of prevention science on how prevention findings can be used to inform epidemiology as it moves along the research continuum from basic science to implementation.

Qualitative methodologies may play an important role in informing the feedback loop between prevention science and epidemiology. Although the use of qualitative methods in the fields of prevention science and epidemiology has been minimal, prevention scientists have developed an increased awareness with respect to the use of qualitative methods, particularly to understanding processes at the implementation stage (Bucher Della Torre, Akré, & Suris, 2010; Pontin, Peters, Lobban, Rogers, & Morriss, 2009; Voogdt-Pruis, Beusmans, Gorgels, & van Ree, 2011). To this end, qualitative methods have great utility in elucidating processes that facilitate or hinder adoption of effective programs, including feasibility and acceptability of prevention strategies among individuals and communities (Parra-Cardona, Cordova, Holtrop, Villarruel, & Wieling, 2008; Parra Cardona et al., 2009), contextual challenges experienced by underserved populations that could serve as barriers to participation (Cervantes & Cordova, 2011; Cordova & Cervantes, 2010), and adaptations to ensure adoption of prevention services in communities (Dodson et al., 2009). Thus, qualitative methodology, when used as a tool to gather prevention findings and process data, has the potential to serve as a feedback loop and inform epidemiology.

To summarize, the field of prevention science could benefit from using prevention findings in the advancement of epidemiologic models. Doing so may prove helpful in transporting a prevention program from bench to practice. As shown in Fig. 1.1, the prevention research cycle should include a feedback loop consisting of multiple steps, including the identification of a problem and large-scale implementation.

Epidemiology and Prevention Science: Working Toward a Mutually Informed Process

In summary, epidemiology plays a significant role in the field of prevention science. Central to epidemiology is the description of the natural history or career of the outcome of interest and consequently focuses on factors that contribute to the onset, progression/escalation, and cessation/desistance of the outcome of interest. Epidemiology provides a framework for prevention scientists to identify the outcome of interest and develop etiologic models and preventive interventions, methods, and study designs to test preventive interventions. Such efforts are aimed at promoting health and well-being in individuals, families, and communities alike. Although a substantial amount of research has demonstrated the ways in which epidemiological findings are useful to the field of prevention science, little has been done with regard to prevention findings informing epidemiologic models. Working toward a mutually informed process will advance the fields of epidemiology and prevention science. A mutually informed process, whereby both epidemiology and prevention science inform each other, may aid in the development of optimally efficacious and effective preventive interventions that promote health and prevent/reduce health-risk behaviors.

Acknowledgement

The preparation of this chapter was supported by National Institute on Drug Abuse (NIDA) grant # R01DA025192-02S1, NIDA and National Institute on Alcohol Abuse and Alcoholism grant #3R01DA025192-01A1S1 awarded to Guillermo Prado, and National Institute on Minority Health and Health Disparities grant #1L60MD006269-01 awarded to David Cordova.

References

Behrendt, S., Wittchen, H. U., Höfler, M., Lieb, R., & Beesdo, K. (2009). Transitions from first substance use to substance use disorders in adolescence: Is early onset associated with a rapid escalation? Drug and Alcohol Dependence, 99(1–3), 68–78.PubMed

Benson, K., & Hartz, A. J. (2000). A comparison of observational studies and randomized, controlled trials. New England Journal of Medicine, 342(25), 1878–1886. doi:10.​1056/​NEJM200006223422​506.PubMed

Berrueta-Clement, J. R., Schweinhart, L. J., Barnett, W. S., & Weikart, D. P. (1987). The effects of early educational intervention on crime and delinquency in adolescence and early adulthood. In J. D. Burchard & S. N. Burchard (Eds.), Prevention of delinquent behavior (Vol. X, pp. 220–240). Newbury Park, CA: Sage Publications.

Blackburn, A. C., & Jerry, D. J. (2011). Map making in the 21st century: Charting breast cancer susceptibility pathways in rodent models. Journal of Mammary Gland Biology and Neoplasia, 16, 57–64.PubMed

Botvin, G. J., Griffin, K. W., Paul, E., & Macaulay, A. P. (2003). Preventing tobacco and alcohol use among elementary school students through life skills training. Journal of Child and Adolescent Substance Abuse, 12(4), 1–7. doi:10.​1300/​J029v12n04_​01.

Braveman, P., & Barclay, C. (2009). Health disparities beginning in childhood: A life-course perspective. Pediatrics, 124(Suppl 3), S163–S175. doi:10.​1542/​peds.​2009-1100D.PubMed

Brody, G. H., Beach, S. R., Philibert, R. A., Chen, Y. F., & Murry, V. M. (2009). Prevention effects moderate the association of 5-HTTLPR and youth risk behavior initiation: Gene x environment hypotheses tested via a randomized prevention design. Child Development, 80(3), 645–661. doi:10.​1111/​j.​1467-8624.​2009.​01288.​x.PubMed

Brody, G. H., Chen, Y. F., Beach, S. R., Philibert, R. A., & Kogan, S. M. (2009). Participation in a family-centered prevention program decreases genetic risk for adolescents’ risky behaviors. Pediatrics, 124(3), 911–917. doi:10.​1542/​peds.​2008-3464.PubMedCentralPubMed

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Annals of child development (Vol. 6, pp. 18–249). Greenwich, CT: JAI Press.

Brown, C. H., & Faraone, S. V. (2004). Prevention of schizophrenia and psychotic behavior: Definitions and methodologic issues. In W. S. Stone, S. V. Faraone, & M. T. Tsuang (Eds.), Early clinical intervention and prevention in schizophrenia (pp. 255–284). Totowa, NJ: The Humana Press.

Brown, C. H., Ten Have, T. R., Jo, B., Dagne, G., Wyman, P. A., Muthén, B., et al. (2009). Adaptive designs for randomized trials in public health. Annual Review of Public Health, 30(1), 1–25. doi:10.​1146/​annurev.​publhealth.​031308.​100223.PubMedCentralPubMed

Brown, C. H., Wang, W., Kellam, S. G., Muthén, B. O., Petras, H., Toyinbo, P., et al. (2008). Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and time. Drug and Alcohol Dependence, 95(Suppl 1), S74–S104. doi:10.​1016/​j.​drugalcdep.​2007.​11.​013.PubMedCentralPubMed

Brown, C. H., Wyman, P. A., Guo, J., & Peña, J. (2006). Dynamic wait-listed designs for randomized trials: New designs for prevention of youth suicide. Clinical Trials, 3(3), 259–271. doi:10.​1191/​1740774506cn152o​a.PubMed

Brug, J., Oenema, A., & Ferreira, I. (2005). Theory, evidence and intervention mapping to improve behavior nutrition and physical activity interventions. International Journal of Behavioral Nutrition and Physical Activity, 2(1), 2.PubMedCentralPubMed

Bucher Della Torre, S., Akré, C., & Suris, J. C. (2010). Obesity prevention opinions of school stakeholders: A qualitative study. Journal of School Health, 80(5), 233–239. doi:10.​1111/​j.​1746-1561.​2010.​00495.​x.PubMed

Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of antisocial behavior. In J. D. Hawkins (Ed.), Delinquency and crime: Current theories (pp. 149–197). New York, NY: Cambridge University Press.

Cervantes, R. C., & Cordova, D. (2011). Life experiences of Hispanic adolescents: Developmental and language considerations in acculturation stress. Journal of Community Psychology, 39(3), 336–352. doi:10.​1002/​jcop.​20436.

Chambers, R. A., Taylor, J. R., & Potenze, M. N. (2003). Developmental neurocircuitry of motivation in adolescence: A critical period of addiction vulnerability. American Journal of Psychiatry, 160(6), 1041–1052.PubMedCentralPubMed

Cleveland, M. J., Feinberg, M. E., & Greenberg, M. T. (2010). Protective families in high- and low-risk environments: Implications for adolescent substance use. Journal of Youth and Adolescence, 39(2), 114–126. doi:10.​1007/​s10964-009-9395-y.PubMedCentralPubMed

Cordova, D., & Cervantes, R. C. (2010). Intergroup and within-group perceived discrimination among U.S.-born and foreign-born Latino youth. Hispanic Journal of Behavior Science, 32(2), 259–274. doi:10.​1177/​0739986310362371​.

Cordova, D., Huang, S., Arzon, M., Freitas, D., Malcolm, S., & Prado, G. (2011). The role of attitudes, family, peer and school on alcohol use, rule breaking and aggressive behavior in Hispanic delinquent adolescents. Open Family Studies Journal, 4, 38–45.PubMedCentralPubMed

DiClemente, R. J., Wingood, G. M., Harrington, K. F., Lang, D. L., Davies, S. L., Hook, E. W., 3rd, et al. (2004). Efficacy of an HIV prevention intervention for African American adolescent girls. JAMA, 292(2), 171–179. doi:10.​1001/​jama.​292.​2.​171.PubMed

Dishion, T. J., Kavanagh, K., Schneiger, A., Nelson, S., & Kaufman, N. K. (2002). Preventing early adolescent substance use: A family-centered strategy for the public middle school. Prevention Science, 3(3), 191–201.PubMed

Dodson, E. A., Fleming, C., Boehmer, T. K., Haire-Joshu, D., Luke, D. A., & Brownson, R. C. (2009). Preventing childhood obesity through state policy: Qualitative assessment of enablers and barriers. Journal of Public Health Policy, 30(S1), S161–S176.PubMed

Donelan-McCall, N., Eckenrode, J., & Olds, D. L. (2009). Home visiting for the prevention of child maltreatment: Lessons learned during the past 20 years. Pediatric Clinics of North America, 56(2), 389–403. doi:10.​1016/​j.​pcl.​2009.​01.​002.PubMed

Duncan, T. E., Duncan, S. C., & Strycker, L. A. (2006). An introduction to latent variable growth curve modeling: Concepts, issues, and applications. Mahwah, NJ: Lawrence Erlbaum Associates. Revised Edition.

Finkelstein, M. O., Levin, B., & Robbins, H. (1996a). Clinical and prophylactic trials with assured new treatment for those at greater risk: I, A design proposal. American Journal of Public Health, 86(5), 691–695.PubMedCentralPubMed

Finkelstein, M. O., Levin, B., & Robbins, H. (1996b). Clinical and prophylactic trials with assured new treatment for those at greater risk: II, Examples. American Journal of Public Health, 86(5), 696–705. doi:10.​2105/​ajph.​86.​5.​696.PubMedCentralPubMed

Franks, P. W., & Poveda, A. (2011). Gene-lifestyle and gene-pharmacotherapy interactions in obesity and its cardiovascular consequences. Current Vascular Pharmacology, 9(4), 401–456.PubMed

Glasgow, R. E., McKay, H. G., Piette, J. D., & Reynolds, K. D. (2001). The RE-AIM framework for evaluating interventions: What can it tell us about approaches to chronic illness management? Patient Education and Counseling, 44(2), 119–127. doi:10.​1016/​s0738-3991(00)00186-5.PubMed

Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89(9), 1322–1327. doi:10.​2105/​ajph.​89.​9.​1322.PubMedCentralPubMed

Gordis, L. (2009). Epidemiology (4th ed.). Philadelphia, PA: Saunders.

Grant, B. F., & Dawson, D. A. (1998). Age at onset of drug use and its association with DSM–IV drug abuse and dependence: Results from the National Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 10, 163–173.PubMed

Griffin, K. W., & Botvin, G. J. (2010). Evidence-based interventions for preventing substance use disorders in adolescents. Child and Adolescent Psychiatric Clinics of North America, 19(3), 505–526. doi:10.​1016/​j.​chc.​2010.​03.​005.PubMedCentralPubMed

Griffin, K. W., Botvin, G. J., Nichols, T. R., & Doyle, M. M. (2003). Effectiveness of a universal drug abuse prevention approach for youth at high risk for substance use initiation. Preventive Medicine, 36(1), 1–7.PubMed

Grimshaw, J., Campbell, M., Eccles, M., & Steen, N. (2000). Experimental and quasi-experimental designs for evaluating guideline implementation strategies. Family Practice, 17(Suppl 1), S11–S16.PubMed

Guarino, H., Deren, S., Mino, M., Sung-Yeon, K., & Shedlin, M. G. (2010). Training drug treatment patients to conduct peer-based HIV outreach: An ethnographic perspective on peers’ experiences. Substance Use & Misuse, 45(3), 414–436.

Harper, G. W., Bangi, A. K., Sanchez, B., Doll, M., & Pedraza, A. (2009). A quasi-experimental evaluation of a community-based HIV prevention intervention for Mexican American female adolescents: The SHERO’s program. AIDS Education and Prevention, 21, 109–123.PubMed

Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105. doi:10.​1037/​0033-2909.​112.​1.​64.PubMed

Hawkins, J. D., Oesterle, S., Brown, E. C., Monahan, K. C., Abbott, R. D., Arthur, M. W., et al. (2012). Sustained decreases in risk exposure and youth problem behaviors after installation of the Communities That Care prevention system in a randomized trial. Archives of Pediatrics & Adolescent Medicine, 166(2), 141–148. doi:10.​1001/​archpediatrics.​2011.​183.

Hayes, A. (2006). Maintaining the gains sustainability in prevention and early intervention. Family Matters, 75, 66–69.

Heath, A. C., Bucholz, K. K., Madden, P. A., Dinwiddie, S. H., Slutske, W. S., Bierut, L. J., et al. (1997). Genetic and environmental contributions to alcohol dependence risk in a national twin sample: Consistency of findings in women and men. Psychological Medicine, 27(6), 1381–1396.PubMed

Helbig, K. L., Nothnagel, M., Hampe, J., Balschun, T., Nikolaus, S., Schreiber, S., et al. (2012). A case-only study of gene-environment interaction between genetic susceptibility variants in NOD2 and cigarette smoking in Crohn’s disease aetiology. BMC Medical Genetics, 113, 14. doi:10.​1186/​1471-2350-13-14.

Hemingway, H., & Marmot, M. (1999). Psychosocial factors in the aetiology and prognosis of coronary heart disease: Systematic review of prospective cohort studies. BMJ, 318(7196), 1460–1467. doi:10.​1136/​bmj.​318.​7196.​1460.PubMedCentralPubMed

Hemphill, S. A., Heerde, J. A., Herrenkohl, T. I., Patton, G. C., Toumbourou, J. W., & Catalano, R. F. (2011). Risk and protective factors for adolescent substance use in Washington state, the United States and Victoria, Australia: A longitudinal study. Journal of Adolescent Health, 49(3), 312–320. doi:10.​1016/​j.​jadohealth.​2010.​12.​017.PubMedCentralPubMed

Hops, H., Ozechowski, T. J., Waldron, H. B., Davis, B., Turner, C. W., Brody, J. L., et al. (2011). Adolescent health-risk sexual behaviors: Effects of a drug abuse intervention. AIDS and Behavior, 15(8), 1664–1676.PubMedCentralPubMed

Howe, G. W., Beach, S. R., & Brody, G. H. (2010). Microtrial methods for translating gene-environment dynamics into preventive interventions. Prevention Science, 11(4), 343–354.PubMedCentralPubMed

Ibrahim, J. K., Anderson, E. D., Burris, S. C., & Wagenaar, A. C. (2011). State laws restricting driver use of mobile communications devices: Distracted-driving provisions, 1992–2010. American Journal of Preventive Medicine, 40(6), 659–665. doi:10.​1016/​j.​amepre.​2011.​02.​024.PubMed

Jadad, A. R., Moore, R. A., Carroll, D., Jenkinson, C., Reynolds, D. J., Gavaghan, D. J., et al. (1996). Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Control Clinical Trials, 17(1), 1–12.

Johnston, L. D., O’Malley, P. M., Bachman, J. G., & Schulenberg, J. (2013). Monitoring the future national results on drug use: 2012 Overview, key findings on adolescent drug use. Ann Arbor, MI: Institute for Social Research, The University of Michigan.

Kaufman, J., Yang, B. Z., Douglas-Palumberi, H., Crouse-Artus, M., Lipschitz, D., Krystal, J. H., et al. (2007). Genetic and environmental predictors of early alcohol use. Biological Psychiatry, 61(11), 1228–1234. doi:10.​1016/​j.​biopsych.​2006.​06.​039.PubMed

Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., Toyinbo, P., et al. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, psychiatric, and social outcomes. Drug and Alcohol Dependence, 95(Suppl 1), S5–S28. doi:10.​1016/​j.​drugalcdep.​2008.​01.​004.PubMedCentralPubMed

Kellam, S. G., & Langevin, D. J. (2003). A framework for understanding evidence in prevention research and programs. Prevention Science, 4(3), 137–153. doi:10.​1023/​a:​1024693321963.PubMed

Kellam, S. G., Mackenzie, A. C., Brown, C. H., Poduska, J. M., Wang, W., Petras, H., et al. (2011). The good behavior game and the future of prevention and treatment. Addiction Science & Clinical Practice, 6(1), 73–84.

Kellam, S. G., & Van Horm, Y. V. (1997). Life course development, community epidemiology, and preventive trials: A scientific structure for prevention research. American Journal of Community Psychology, 25(4), 177–188.PubMed

Kellam, S. G., Werthamer-Larsson, L., Dolan, L. J., Brown, C. H., Mayer, L. S., Rebok, G. W., et al. (1991). Developmental epidemiologically based preventive trials: Baseline modeling of early target behaviors and depressive symptoms. American Journal of Community Psychology, 19(4), 563–584.PubMed

Kuh, D., Ben-Shlomo, Y., Lynch, J., Hallqvist, J., & Power, C. (2003). Life course epidemiology. Journal of Epidemiology and Community Health, 57(10), 778–783.PubMedCentralPubMed

Kulis, S., Marsiglia, F., Ayers, S., Calderón-Tena, C., & Nuño-Gutiérrez, B. (2011). Gender differences in drug resistance skills of youth in Guanajuato, Mexico. Journal of Primary Prevention, 32(2), 113–127. doi:10.​1007/​s10935-011-0239-7.PubMedCentralPubMed

Kulis, S., Yabiku, S. T., Marsiglia, F. F., Nieri, T., & Crossman, A. (2007). Differences by gender, ethnicity, and acculturation in the efficacy of the Keepin’ it REAL model prevention program. Journal of Drug Education, 37(2), 123–144.PubMedCentralPubMed

MacKinnon, D. P., & Luecken, L. J. (2008). How and for whom? Mediation and moderation in health psychology. Health Psychology, 27(2), S99–S100. doi:10.​1037/​0278-6133.​27.​2(Suppl.​).​S99.PubMedCentralPubMed

Martinez, C. R., Jr., & Eddy, J. M. (2005). Effects of culturally adapted parent management training on Latino youth behavioral health outcomes. Journal of Consulting and Clinical Psychology, 73(5), 841–851.PubMed

McArdle, J. J. (2009). Latent variable modeling of differences and changes with longitudinal data. Annual Review of Psychology, 60(1), 577–605. doi:10.​1146/​annurev.​psych.​60.​110707.​163612.PubMed

Muthen, L. K., & Muthen, B. O. (1998–2010). Mplus user’s guide. Los Angeles, CA: Author.

National Highway Traffic Safety Administration. (2009). An examination of driver distraction as recorded in NHTSA databases.

Nkansah-Amankra, S., Diedhiou, A., Agbanu, H., Harrod, C., & Dhawan, A. (2011). Correlates of sexual risk behaviors among high school students in Colorado: Analysis and implications for school-based HIV/AIDS programs. Maternal and Child Health Journal, 15(6), 730–741. doi:10.​1007/​s10995-010-0634-3.PubMed

Olds, D. L. (2008). Preventing child maltreatment and crime with prenatal and infancy support of parents: The nurse-family partnership. Journal of Scandinavian Studies in Criminology and Crime Prevention, 9(Suppl 1), 2–24. doi:10.​1080/​1404385080245009​6.PubMedCentralPubMed

Olds, D. L., Henderson, C. R., & Kitzman, H. (1994). Does prenatal and infancy nurse home visitation have enduring effects on qualities of parental caregiving and child health at 25 to 50 months of life? Pediatrics, 93(1), 89–98.PubMed

Olds, D. L., Henderson, C. R., Kitzman, H., & Cole, R. (1995). Effects of prenatal and infancy nurse home visitation on surveillance of child maltreatment. Pediatrics, 95(3), 365–372.PubMed

Olsen, J., Christensen, K., Murray, J., & Ekbom, A. (2010). The randomized controlled trial (RCT): An introduction to epidemiology for health professionals (Vol. 1, pp. 81–84). New York, NY: Springer.

Pantin, H., Coatsworth, J. D., Feaster, D. J., Newman, F. L., Briones, E., Prado, G., et al. (2003). Familias Unidas: The efficacy of an intervention to promote parental investment in Hispanic immigrant families. Prevention Science, 4(3), 189–201.PubMed

Pantin, H., Prado, G., Lopez, B., Huang, S., Tapia, M. I., Schwartz, S. J., et al. (2009). A randomized controlled trial of Familias Unidas for Hispanic adolescents with behavior problems. Psychosomatic Medicine, 71(9), 987–995. doi:10.​1097/​PSY.​0b013e3181bb2913​.PubMedCentralPubMed

Parra Cardona, J., Holtrop, K., Córdova, J. D., Escobar-Chew, A. R., Horsford, S., Tams, L., et al. (2009). Queremos Aprender: Latino immigrants’ call to integrate cultural adaptation with best practice knowledge in a parenting intervention. Family Process, 48(2), 211–231. doi:10.​1111/​j.​1545-5300.​2009.​01278.​x.

Parra-Cardona, J. R., Cordova, D., Holtrop, K., Villarruel, F. A., & Wieling, E. (2008). Shared ancestry, evolving stories: Similar and contrasting life experiences described by foreign born and U.S. born Latino parents. Family Process, 47(2), 157–172. doi:10.​1111/​j.​1545-5300.​2008.​00246.​x.PubMed

Pontin, E., Peters, S., Lobban, F., Rogers, A., & Morriss, R. (2009). Enhanced relapse prevention for bipolar