Sheldon Cohen - US grants

The aims of the proposed research are to elucidate the psychosocial factors critical to maintenance or relapse in self-quitting of smoking and to evaluate the outcomes of some prototypical ways by which smokers try to quit without the help of formal programs. The psychosocial factors to be examined will include stressful life events, perceived stress social support social network characteristics (e.g., interactions with other smokers) and coping strategies. The covariation (and interaction) of these factors with relapse or maintenance will be compared in four samples of subjects: New Years resolution quitters, participants in two different instruction manual programs, and an instruction manual plus four optional group sessions program. Results from previous work relating these psychosocial factors to clinic aided quitting will also be compared with the self-quitting data. Follow-up smoking status and some probes for the psychosocial factors will be obtained at one two three, six and twelve months after the initial quit attempt. Assessments will be carried out by telephone interview. Multivariate procedures, e.g., stepwise multiple regression, will be used to evaluate the relation of psychosocial factors to outcome. The reactivity of the assessment procedures will be experimentally evaluated by comparing groups which receive minimal assessment with those receiving the entire assessment battery. Both the outcomes and cost effectiveness of the four programs will be evaluated. It is expected that the programs providing more assistance (e.g. group sessions) will be more successful but less cost effective than programs providing minimal assistance.

The primary purpose of the proposed research is to determine the role of natural social support systems in individual susceptibility to respiratory infection, and related symptomatic behavior. Alternative conceptual measures of social support and psychosocial stress will be used to predict the incidence and severity of colds for volunteers subsequently inoculated with viruses. We focus on distinguishing between two models of the support-illness relationship: the buffering model--social support reduces illness by protecting persons from the pathogenic effects of stressful events, and the main-effect model--support reduces illness irrespective of stress level. We also examine the roles of positive and negative affect, self-esteem, and personal control in mediating the relationship between social support and illness. By employing a prospective design in which healthy (unifected) subjects are exposed to cold or influenza viruses, we are able to focus on the onset of infection, eliminate the possibility that support results in selective exposure to the agent, and reduce the probability that illness or illness behavior could be responsible for shifts in social networks or perceptions of social support. Random assignment of volunteers to virus or placebo also allows us to determine the psychosocial predictors of symptomatic behavior holding infection constant. Primary outcomes include infection (virus shedding), physical symptoms (quantify of nasal secretion) and behavioral symptoms (symptom reporting, use of handkerchiefs).

This is a request for an ADAMHA RSDA Level II. During the fourteen years since receiving my Ph.D., my research has focused on the identification of pathways through which environmental and psychosocial stressors, and social supports influence behavior, psychological distress and physical health. My current work addresses the possible influences of stress and support on the etiology of disease. The goal of this work is to develop and test sophisticated psycho-biologic models of the etiologic role of psychosocial factors in disease pathogenesis; in particular in relation to upper respiratory infection (URI), and to coronary heart disease (CHD). The purpose of the proposed award is to allow sufficient time to accomplish several goals: (a) the acquisition of additional knowledge in regard to the biologic processes that may mediate the links between psychosocial factors and URI, and psychosocial factors and CHD, (b) the development of sophisticated models of the psycho-biologic pathways linking stress and support to each of these diseases, (c) the pursuit of research goals in regard to testing the URI models, and ultimately the CHD models, and (d) the acquisition of statistical skills necessary to appropriately test the validity of hypothetical psycho-biologic pathways. The proposed empirical work is a prospective study of the influences of stress and social support on susceptibility to URI. A variety of measures of stress and social support are collected from healthy subjects who are subsequently inoculated with a cold or influenza virus (or placebo). The primary outcomes are assessed for four days following inoculation and include: infection (virus shedding), physical symptoms (quantity of nasal secretion), and behavioral symptoms (symptom reporting, use of handkerchiefs). The combination of a prospective design and the manipulation of the infectious agent allow us to focus on the onset of infection, eliminate the possibility that stress or support result in selective exposure to the infectious agent, and reduce the probability that association are attributable to the illness causing shifts in stress or support. Preinoculation measures of health behaviors, psychologic states, and immune status, and immune-response to the viral challenge, are assessed as indicators of possible pathways through which psychosocial factors influence infection and symptomatology.

This is a proposal to continue a research program in which the primary goal is to extend our understanding of the role of psychosocial factors in susceptibility to infectious disease. The major focus of the proposed study is to evaluate the roles of social networks 1 interactions, and interpersonal conflicts and strains in susceptibility to upper respiratory infections. It also attempts to identify causal pathways (psychological, health practice and biological) linking these factors to disease susceptibility. Before being exposed to an upper respiratory virus, 320 volunteers are administered a range of measures to assess social characteristics thought to contribute to or result from social network integration (e.g., social network structure, quantity and quality of day-to-day interactions, interpersonal dispositions, chronic social strains) as well as measures of pathways thought to link social integration to disease susceptibility. Psychological characteristics assessed as possible links between social integration factors and host susceptibility include self-esteem, personal control, and positive and negative affect. Health practices include smoking, alcohol consumption, sleep, diet, and exercise; and endocrine measures include basal levels of epinephrine (epi), norepinephrine (norepi) and cortisol, and the shape of the diurnal cortisol rhythm. Assessment procedures include semi-structured interviews (social support, stress, self- esteem), questionnaires (social networks, control, self-esteem, personality, health practices), a 14-day daily diary (social interactions, affect, health practices), three 24-hour urines (epi and norepi), and 3 days of repeated collection of saliva samples (cortisol). After the psychological, biological, and behavioral assessment procedures are completed, subjects are exposed by nasal drops to one of two rhinoviruses, quarantined and monitored for infection and symptoms for five days. We collect symptom interviews, nasal washings for detection of viral shedding, and objective measures of pathophysiology (temperature, mucus weights, nasal clearance, nasal congestion, presence of inflammatory mediators) before viral-challenge and on each of the 5 days after challenge. Analyses focus on whether interpersonal factors predict clinical illness (virus shedding or four-fold increase in viral- specific antibody plus clinical diagnosis) and whether these relations are explicable in terms of the psychological, endocrine, and health practice mediation.

The primary goal of the research program is to extend our understanding of the role of psychosocial factors in susceptibility to upper respiratory infectious illness. The award allows the investigator (a social psychologist by training) to extend his background in biological, statistical, and methodological areas directly relevant to his research goals. Dr. Cohen has spent his career studying the influence of psychological stress and social supports on mental and physical health. He has made a range of theoretical and empirical contributions and trained over a dozen contributing scientists. With the help of the K02 award, he has received training in basic biological processes and established a research program examining the role of psychosocial factors in infectious disease. He is asking for continued support to expand his expertise in a range of areas relevant to his research plan. The primary aim of this proposed study is to evaluate the roles of social networks, relationships, supports, and interactions in susceptibility to upper respiratory infections. It also attempts to identify causal pathways (psychological, health practice, and endocrine linking these factors to disease susceptibility. Before being intentionally exposed to an upper respiratory virus, 320 volunteers are administered measures of the psychosocial predictors. These factors are assessed with interviews, questionnaires, and daily-diaries. Health-practices, affects, cognitions, and endocrine response are also assessed as potential pathways linking networks and supports to host susceptibility. After these assessment procedures are completed, subjects are exposed by nasal drops to one or two rhinoviruses, quarantined and monitored for infection and symptoms for five days. Analyses focus on whether psychosocial factors predict who develops clinical illness (verified infection plus clinical diagnosis) and whether these relations are explicable in terms of affective, cognitive, endocrine, and health practice mediation. Outcomes based on two definitions of clinical diagnosis, objectively assessed signs of disease, and patient reports of symptoms and disease, are studied and contrasted.

stress; respiratory infections; psychoneuroimmunology; health behavior; leukocyte activation /transformation; cardiovascular system; natural killer cells; personal log /diary; behavioral /social science research tag; human subject; clinical research;

DESCRIPTION (adapted from investigator's abstract): This project represents a continuation of Dr. Cohen's well known studies on the psychobiological factors which influence individuals' susceptibility to the clinical manifestations of viral infections. In this particular series of studies, particular emphasis will be placed on the effects of marital relationships. The study will be a prospective analysis of 160 healthy volunteers, of whom 2/3 will be married, who receive an initial evaluation on hypothesized predictors and mediators of illness, and who will be subsequently inoculated with one of tow viruses that cause a mild influenza illness. The investigators will monitor indicators of infection and illness expression over the next 7 days after inoculation. Despite the fact that 98% will become infected, only about 40% will develop signs of "illness," and the latter will be the main outcome variable. The specific aims are to: (1) identify the characteristics of marital and non-marital relationships that contribute to resistance; (2) determine the influence of personality characteristics; (3) determine the biological pathways by which social relationships influence resistance; and (4) assess the influence of demographic variables such as age, SES, and perceived social status on resistance to infection. The rationale underlying the study is that there is a lot of suggestive but not conclusive evidence that married people are generally healthier than single or formerly married people, but the causal relationships and biological pathways mediating this association is unknown. During the 2 months before the study subjects will undergo extensive evaluation that will include personality and socials environment questionnaires, urine collection for catecholamines, and repeated salivary cortisol samples. For the study itself they will be housed in a hotel, where they will remain for 8 days. At the end of the first day they are inoculated with the virus, and during the ensuing week they are closely monitored for symptoms, measurement of the virus and antibodies, salivary cortisols, and daily affect and health practices. The major independent variable in the analysis will be the quality of the marriage: unmarried, low quality, or high quality. It is hypothesized that a high quality marriage will result in less illness expression.

DESCRIPTION (provided by applicant): Our primary goals are to assess the influence of the social environment on susceptibility to upper respiratory illness and to identify the pathways through which environmental factors "get under the skin" to influence disease expression. We hypothesize that the social environment influences behavior, hypothalamic-pituitary-adrenocortical (HPA) and autonomic nervous system (ANS) responses which in turn influence the regulation of both cellular and humoral immune response to the virus. We use a unique prospective design. We first assess characteristics of the social environment and the status of behavioral and physiological systems we predict to link these characteristics to host resistance in 260 healthy adults who are seronegative for antibodies to the challenge virus. Subsequently, we expose each subject by intranasal drops to one of two strains of rhinovirus and monitor them closely for evidence of infection and illness in a cloistered environment. Measures of illness collected immediately before and daily during the cloister include virus shedding and titer, nasal pro-inflammatory, anti-inflammatory and anti-viral cytokine production, nasal and general symptoms, body temperature, nasal secretion production and nasal mucociliary clearance function. Using these data, we will determine if the social environment influences illness expression by altering baseline cellular immune status (delayed type hypersensitivity), Th-1/Th-2 cytokine balance, and the ability to produce pro-inflammatory cytokines (from mitogen stimulated peripheral blood mononuclear cells); whether it influences the regulation of cytokine production in response to the infection; and whether the association between the social environment and immunity are mediated by the ANS (separating sympathetic and parasympathetic contributions) and/or HPA systems. All analyses will control for demographic factors and will focus on whether the social environment predicts clinical illness (verified infection + clinical diagnosis). The results of this study will define the pathways by which one's social environment influences their susceptibility to upper respiratory virus infection and may identify potential targets for behavioral or pharmaceutical interventions to restore disease resistance in persons most at risk because of their social environments.

DESCRIPTION (provided by applicant): This application addressed Broad Challenge Area: (03) Biomarker Discovery and Validation and specific challenge topic: 03-AT-101* Psychoneuroimmunology biomarkers of stress. Identification of biomarkers to assess the impact of stress, both social and biological, on immune function. The rate of cell senescence in lymphocytes (particularly in CD8 cells), as indicated by decreases in telomere length, may be an important new marker of immunocompetence and, in turn, host resistance to infectious agents. Loss of telomere length may be especially important in understanding the role of psychological stress in infectious illness. Stress is associated with greater susceptibility to upper respiratory viral illnesses. In addition, increases in psychological stress are associated with shorter telomere length in lymphocytes. This evidence suggests that stress may influence disease susceptibility via its effects on lymphocyte cell senescence. Moreover, psychological stress has also been associated with greater oxidative stress and less telomerase activity. Both of these contribute to shortening of telomeres, suggesting that psychological stress may shorten telomeres via its effects on these telomere regulators. We have a unique opportunity to address these hypotheses in healthy adult human subjects by adding assessments of oxidative stress, telomerase activity and telomere length at baseline to an ongoing viral- challenge study. Baseline measures in the parent study include a range of stress measures: psychological and social measures of stress by questionnaire, assessments of basal diurnal cortisol rhythms over three days and of autonomic and cortisol response to an experimental stressor in the laboratory. After completing all baseline measures, subjects are exposed to a safety tested rhinovirus and monitored in quarantine for the development of infection and illness. Local (nasal secretory) release of pro- and anti- inflammatory cytokines are also assessed after viral exposure. Approximately 40% of the subjects develop a clinical illness in response to the viral challenge. In this extension of that study, we propose to evaluate whether telomere length predicts susceptibility to infection and disease expression and whether it constitutes a mediating pathway through which stress influences resistance to infectious disease. We can also determine whether or not telomere length mediates the known effects of psychological stress on the locally produced pro- and anti-inflammatory cytokines that play a role in illness expression. Finally, we can test whether or not telomerase activity and oxidative stress constitute pathways through which psychological, social and biological markers of stress might influence telomere length. This work has implications for understanding the role of stress in the aging process and particularly important implications for reducing disease in the elderly and those experiencing long-term stress (e.g. the chronically unemployed, caregivers, and those suffering from poverty). PUBLIC HEALTH RELEVANCE: A major focus of the project is to determine whether immune cell senescence (as assessed by telomere length) plays an important role in our ability to fight off infection. Because life stress has been found to be associated with greater senescence, we are also testing whether stress-associated increases in senescence can account for why life stress increases our risk for infectious diseases. An in-depth understanding of the role how stress influences our health can help lead to effective interventions to protect people at risk.

DESCRIPTION (provided by applicant): This proposed project plans to produce, document and archive a data repository that combines final research data from 5 prospective viral-challenge studies (9 different viruses) focused on examining the roles of psychosocial factors in susceptibility to the common cold. In these unique studies, hypothesized predictor and mediating variables were assessed in healthy adults aged 18-55 years, who were then experimentally exposed to a virus that causes the common cold and monitored for development of infection and signs and symptoms of illness. Clinical illness (found in 40% of volunteers) was defined as both being infected and expressing signs and symptoms of disease. This design allowed the determination of whether variables of interest in a given study (e.g., stress, personality) predicted who developed a clinical cold following viral exposure. Merging the data from these 5 studies provides the opportunity to replicate analyses across two or more of the studies or combine data from any or all of the studies, allowing adequate power for detection of small effects and sophisticated tests of mediation and moderation. The first aim of this proposal is to use the combined data to address important questions about the role of social relationships in health. Both the quantity and quality of social relationships were assessed at baseline using both questionnaires and daily interviews over 6-14 days. Analysis of the aggregated dataset will be conducted to help identify (1) the types of social ties that matter most for health;(2) the unique mechanisms linking different types of social ties to health;(3) components of social relationships likely to be conducive to intervention;and (4) characteristics of individuals that make them more or less responsive to the influences of social ties. Second, in addition to these research aims, a major thrust of this application is to disseminate the combined data from these 5 studies to the broader scientific community. Although each study was designed to address a specific set of hypotheses, many variables were collected across the multiple studies in up to 1,400 participants. These included: standard control variables/covariates (e.g., age, sex, race, body mass, season, and specific antibody titer to the challenge virus);predictor and mediator variables such as measures of social relationships, stress, personality, affect, socioeconomic status, health practices, local cytokine response, and endocrine levels;and assessment of illness (colds) including viral-related biology and symptom/sign outcomes. Other outcomes common across studies included basal blood pressure, heart rate, and pulmonary function. Individual studies also included markers of biological aging and assessments of acute stress response. These unique and valuable data could be used to address numerous questions beyond those included in our research aims. Creating a centralized and publicly available online data repository will have the immediate effect of maximizing their worth to the scientific community. In the long term, findings derived from these data may provide insight into possible alternative targets for intervention that typically are not addressed using conventional medical techniques. PUBLIC HEALTH RELEVANCE: The proposed project plans to combine, analyze and make publicly available data from 5 viral-challenge studies focused on examining the roles of psychological and social factors in susceptibility to the common cold. Our specific interest is in the role of social relationships in the development of signs and symptoms of disease. Because all 5 studies collected data on multiple psychosocial, behavioral, and biological variables, the public availability of these findings will provide investigators from various disciplines with the opportunity to examine associations of other psychosocial and behavioral factors with multiple measures of disease and disease risk, thus providing insight into possible alternative targets for intervention that typically are not addressed using conventional medical techniques.