摘要：Addiction represents one of the greatest public health problems facing the United States. Advances in addiction research have focused on the neurobiology of this disease. We discuss potential new breakthroughs in understanding the other side of gene-environment interactions—the environmental context or “exposome” of addiction. Such research has recently been made possible by advances in geospatial technologies together with new mobile and sensor computing platforms. These advances have fostered interdisciplinary collaborations focusing on the intersection of environment and behavior in addiction research. Although issues of privacy protection for study participants remain, these advances could potentially improve our understanding of initiation of drug use and relapse and help develop innovative technology-based interventions to improve treatment and continuing care services. IN THE UNITED STATES, MORE than 22 million Americans meet the criteria for substance abuse or dependence. 1 Addiction involving illicit and prescription drugs, tobacco, and alcohol afflicts 40.3 million Americans (15.9% of the population aged 12 years and older), a greater number than those who have heart disease, diabetes, or cancer. 2 We discuss the potential for new breakthroughs in addiction research based on advances in geospatial technologies, and their combination with sensor and mobile technologies, for understanding the environmental context of substance abuse. Relevant geospatial technologies include global positioning systems (GPS), which support the collection of location data in real time, and geographic information systems (GIS), which can be considered a type of spatial database software for acquiring, integrating, analyzing, and visualizing location-based data. Related technologies include sensors that allow for the collection of biometric and environmental exposure data, and mobile computing platforms such as smartphones that can deliver surveys to capture information on behaviors and emotional states as they occur. The academic field of Geographic Information Science focuses on the development and application of these technologies, and integrates components of computer science, engineering, geography, and statistics to investigate how place and environment shape behavior. Recently, nontraditional interdisciplinary collaborations between health researchers and geographic information scientists utilizing geospatial and related technologies have led to a more comprehensive understanding of the interaction between the individual and the environment relating to drug use. This relatively new area of research expands the traditional biopsychosocial concept that has been used to guide psychiatric services and mental health research on addiction to include an analysis of relevant geographic variables to help predict risk factors for the increased incidence of substance use disorders and relapse following successful treatment. Major advances in genomics, molecular research, and brain imaging studies have provided evidence that addiction results from a complex interaction between genes and environment, and recent studies have greatly enhanced our understanding of the etiology, vulnerability, behavioral changes, and neurobiology of this disorder. 3 What is relatively less understood, however, is the other side of the gene–environment interaction (epigenetics)—the environmental context of addiction—that influences initiation into drug use, treatment outcomes, and relapse. We note that environment can be conceptualized as not only the physical environment, but also the social environment (e.g., one’s interactions with other drug users) and the community environment (e.g., a community’s tolerance of drug use). Exposure to addictive substances and the resulting changes in the brain occur within an environmental context that may be termed the “exposome” of addiction—the critical environmental exposures to various stimuli, both physical and social, that interact with one’s genetic and psychological vulnerability to addiction. Christopher Wild, a cancer researcher, coined the term exposome to refer to the totality of environmental exposures throughout the life course as a way of better understanding the environmental causes of disease. 4 Although the types of environmental exposures that lead to addiction are different from those related to cancer, it can be viewed as similar in concept. Addiction occurs within an environmental context, and memories of the stimuli associated with the drug and the context in which use occurred are stored in the brain. These stored memories of the people, paraphernalia, emotions, specific drugs ingested, activities, and physical environment associated with drug use (similar to the well known Alcoholics Anonymous phrase of “people, places, and things”) act as important triggers for cravings that may occur even after long periods of nonuse, especially under conditions of stress and dysphoria. Drug use initiation and subsequent relapse after periods of abstinence are a complex phenomenon, affected by numerous biopsychosocial variables. Although characteristics such as stress, family dynamics, and other personal factors have long been known to influence drug use, the interaction of individual psychological states and genetic vulnerabilities with more macrolevel environmental and geographic factors have only recently been recognized as important influences on addiction. Adolescents may be at greater risk for initiating drug use as a result of a complex interplay between individual psychological and social factors, combined with living in and spending time in certain places that can be categorized as risky or safe environments. 5 In addition, although advances in technology may lead to innovative solutions to address the problem of addiction, they may also contribute to increasing the risk of drug use. For example, because increased availability and accessibility of illicit substances raise the likelihood of drug use, the ability to acquire addictive substances via the Internet and enhanced communications through cellphone and social networking Web sites can facilitate the distribution of drugs and the ease of contact between drug seller and consumer. Thus, in this case, advances in technology have arguably resulted in increasing the exposome of illicit drugs for most people. Although treatment of addictive disorders has, in general, demonstrated considerable efficacy and effectiveness, one of the great challenges to health care providers is the maintenance of long-term abstinence given that addiction is a chronic disease, similar to other disorders with behavioral components, such as diabetes, asthma, and hypertension. 6 Research has established the critical role of the environmental context of drug use in laboratory studies, and how certain kinds of drug-related cue exposures may trigger craving and addictive behaviors. 7 Outside the laboratory, research has also found that where someone lives, and the individual’s physical and social environment, also play an important role in addiction. It is not just an individual’s genetic vulnerability and exposure to addictive substances that results in an addictive disorder. Rather, these etiologic factors occur within a physical and social environmental context that influences the nature of the disease process, as well as one’s likelihood of sustaining recovery after treatment. For example, such neighborhood characteristics as concentrated disadvantage and economic deprivation, a community’s collective efficacy, proximity to alcohol outlets, the amount of abandoned housing nearby, levels of crime, and one’s immediate social environment of family and peer associations have all been shown to influence substance use and recovery. 8–13 However, our current state of knowledge concerning the role of the physical and social environment in addiction has been at a relatively gross and unspecified level of understanding. Although the idea that the environment influences behavior is well established, until very recently, the data and tools necessary to investigate this relationship in a thorough and systematic way were not available. The use of GIS and spatial data has grown so that aggregate data over geographic areas can be easily integrated and analyzed in studies of behavior. However, what is unique to the present is the technological ability to gather spatially and temporally referenced data for individuals—where and when behaviors occur. The development of GPS and its embedding within mobile devices such as smartphones allows for the collection and even streaming of time-stamped location data so that an individual’s daily movements may be tracked at a very high spatial and temporal resolution throughout their day. Ecological momentary assessment (EMA) uses mobile devices to collect data on emotional, social, biologic, and behavioral states and activities in real time as they occur, and in combination with GPS technology, where they occur. 14–16 With current technologies, EMA data can now be collected on the fly via smartphones, wearable sensors, and other mobile computing devices. 17 Combining the technological advances of smartphones, GPS, and EMA makes the ability to gather real-time, spatiotemporally referenced, emotional, and behavioral data possible, which can then lead to a deeper understanding of the factors associated with initiation of drug use as well as relapse. These types of EMA studies offer the potential for examining the individual’s exposure to the environment in a way not possible to do in previous years. Although EMA has been utilized in many studies involving smoking, 18 only relatively recently have EMA studies been conducted with users of illicit drugs. 19 Research is just beginning to examine the relationship between individual states (e.g., mood) collected via EMA combined with geospatial data relating to the environmental context of illicit drug use. 20 The utility of data related to individuals’ space-time paths (i.e., where someone goes and when), and their emotional, social, biologic, and behavioral states and activities can be combined with other data about the natural, built, and social environment to deepen our understanding of drug use initiation, addiction, and relapse in the real world. Thus, it is not only the technologies that facilitate individual-level data collection that are important, but the proliferation and dissemination of digital, spatial data in general. Data vendors in the United States and many other countries now provide spatial and related data on nearly all legal businesses, places of worship, crime, parks, and other features in the environment. The methods associated with collecting data on individuals, however, raises new questions about privacy and maintaining civil liberties. Because data can now be collected on space-time paths of individuals, in addition to their emotions and behaviors, new safeguards must be adopted to protect this information. Conventional methods of privacy protection, such as deidentification of records, may be inadequate when spatial and temporal data on activities are involved. Although researchers have used a variety of strategies to protect the privacy of individuals’ home addresses, including spatial randomization and spatial data aggregation, such strategies may be inadequate because individuals’ identities can often be inferred indirectly from data that express not only an individual’s home location but also the locations of their friends’ and relatives’ homes, places of leisure or religious worship, and other routine activities. With improved safeguards concerning privacy, this newly evolving understanding of the addiction exposome can result in enhanced prevention as well as posttreatment aftercare planning. As mentioned earlier, we are just now beginning to understand that certain environments may be considered “protective” and others “risky” in terms of the likelihood of substance initiation and relapse. 21 These protective and risky influences may be moderated by individual characteristics, such as race and gender, and their immediate social contexts, such as family dynamics and peer affiliations. However, the environmental context of addiction should not be seen simply as a simple, direct cause of addiction, but rather as part of a complex constellation of interacting contextual and genetic factors that influence substance use uptake, addiction, treatment effectiveness, and long-term recovery. Most importantly, there is the potential to harness this technology to not only enhance our understanding of drug use initiation and the precursors to relapse, but to also provide immediate cost-effective, real-time brief interventions through mobile technology. This technology can be potentially used to deliver interventions to prevent impending drug use as it is about to occur, or to provide peer or clinician recovery support anywhere and any time. 22,23 The ability to provide some form of “portable” peer support or even psychiatric intervention to an individual in the neighborhood via a mobile device to prevent relapse is becoming increasingly possible and feasible. Conceivably, interventions could be delivered when sensor- or client-initiated reports of acute stress occur, or when an individual travels to a place where drugs have previously been purchased or used, and therefore, may be identified as risky in terms of relapse potential. Improved understanding of how the physical environment affects and interacts with one’s social environment, genetic vulnerability, and individual characteristics relating to addiction may also lead to improved and healthier physical environments that will help contribute to preventing and improving treatment outcomes for one of the world’s greatest public health problems. The unraveling of the mystery of the human genome has shed light on many critical elements of the disease of addiction, but recent technological advances will enable researchers to better investigate the other side of the gene-environment interaction—the exposome of addiction. Expanding clinical research into the impact of the exposome offers the potential of increasing the capacity to better identify not only at-risk groups, but risky situations and environments. This underscores the value of considering clinical phenotypes from an expanded perspective that is context specific. Additional collaborations among medical, public health, and mental health researchers, and GIS experts are needed to maximize the potential positive impact this area of investigation can have on the future of public health prevention strategies and sustained recovery in the community following treatment. More research that explicitly includes the environmental context of drug use holds the potential for deepening our understanding of addiction, and how community-wide interventions may play an increasingly important role in ameliorating this debilitating major disease.
