Ongoing Collaborative Research

We are continuing to support a number of smaller ongoing collaborative studies in areas of particular relevance to our work. Table 2 provides an overview of ongoing studies and progress to date. These projects provide data on SES, health and mediating pathways across the lifecourse, with special emphasis on children and youth.

TABLE 2 – Ongoing Collaborative Research—financial and/or intellectual support from network
Title of project	PIs/Consultants	Scope of work and progress
Childhood, adolescence, lifecourse SES and longitudinal studies
Rural New York children	Evans	Four wave longitudinal study examining the role of cumulative risk exposure in accounting for greater socioemotional difficulties and elevated physiological stress among low vs. middle-income children in rural New York. The Network has contributed funding at different stages of this project. Currently approximately 50 low- and 50 middle-income, 16-18 year olds who have been studied since 4th grade are being assessed using electronic diaries. Every waking hour for four consecutive days data are gathered on life stressors and hassles, coping strategies, and emotional affect. The third wave of data collection was completed in July, 2006, and analyses are going forward. The fourth wave of data collection has been pilot tested and has just launched. Gary is collaborating with Martha Farah on neurocognitive evaluations of 22 year olds who have been in the rural poverty sample since they were 8 years old. They have written an NIH grant proposal that will be revised and resubmitted to fund this piggy back onto the ongoing project on rural poverty and children’s development.
Early influences of socioeconomic position on the development of the hippocampus: A potential marker for vulnerability to stress and its impact on mental health.	Lupien, Pruessner, Bruce McEwen	Pre-determined individual differences in hippocampal volumes (HV) can set the threshold for vulnerability to stress and its impact on mental health. Exposure to various stimuli early in life (SES, environmental demands and enrichments, stress, etc.) could predict patterns of reactivity to stress, and consequently the development or not of resilience in the face of adversity. This study will examine the relationship between early adversity (e.g., low income, maternal depressive symptomatology) and HV. A longitudinal cohort of normal children followed from birth has been identified in Montreal; they are currently turning 10, the age of choice for the study. Data are available on mother and child depressive symptomatology, and a wealth of psychosocial and SES data are also available. More than three quarters of the children have been scanned. The team is waiting to finish up the entire scanning of the children before starting the volumetric analysis since they want to have all 40 brains in order to ensure rater validity. They should be done with the testing by mid-June and should start analyzing by third week of June.
Development of optimism	Raikkonen, Matthews,Scheier	Focus in utero and early childhood factors that influence the development of optimistic attitudes of children at age 7 and the health correlates of optimism in light of the socioeconomic status of the families. Specific research questions: 1. What are the associations between socioeconomic status and young children's optimistic attitudes and persistent behavior? 2. Do optimistic children exhibit smaller stress responses, especially cortisol responses? 3. Does fetal exposure to black licorice impact the development of optimism, especially in low socioeconomic status families? 312 children have been tested. The mothers of the children reported eating high amounts of licorice (glycyrrhinizin, a natural constituent of licorice, inhibits 11beta-HSD2, which is an enzyme that converts glucocorticoids into their inert form) during pregnancy and 186 child-mother pairs, whose mothers did not report eating high amount of licorice during pregnancy. Measures include detailed questionnaires by the mother, teacher, and child; parental characteristics, neurocognitive assessment of the children, a clinical examination of body size and blood pressure; children’s completion of a psychophysiological study during which salivary cortisol, heart rate, cardiac output, a marker of sympathetic activity, and continuous blood pressure were measured. Children wore an actigraph for 3 days to assess sleep duration and efficiency. Optimism was assessed using a scale which asks children to imagine themselves in a series of events (one at a time), and then asks them to indicate whether the outcome associated with each event (some good and some bad) is likely to occur. A behavioral measure of persistence at difficult problems (adjusted for children’s ability) was included because optimistic adults are known to persist longer than pessimists at challenging tasks. Data collection is complete. The team has met to review the first set of analyses and will continue working together over the next year.
Dynamic models of socioeconomic status and child health	Chen, Martin, Matthews	Empirically test longitudinal relationships between family SES effects and child health. In particular, test that trajectories of SES will impact the health status of adolescents, and determine whether certain periods of SES during childhood are more important for predicting adolescent health outcomes. The project is using NLSY-Children to better understand SES and physical health relationships among 10-11 year olds and 14-15 year olds. Analyses yielded three papers.
St. Louis and Pittsburgh Adolescence Study	Matthews, Chen	Study examining the association between negative interpretations of ambiguous situations and ambulatory blood pressure in the St. Louis and Pittsburgh Adolescence project. At this point the most intriguing finding is that adolescents who make negative interpretations show elevations in nighttime blood pressure and heart rate (in press). In addition, pessimistic attitudes measured via the LOT were related to nighttime systolic blood pressure (papers under review). These findings suggest that there may be a cost to making negative interpretations about others and the future that spills into nighttime physiology.
Pittsburgh high school study	Matthews, McGarth	Study looking at the effect of interpersonal conflict at school on daytime ambulatory blood pressure, blood pressure dipping at night (an indication of good cardiovascular functioning) and carotid artery narrowing. Questionnaires and interviews assess personality, stress, vigilance for threat, and social conflict. Results show that neighborhood level of income but not family income is associated with elevated ambulatory blood pressure and that those who live in poorer neighborhoods show greater increases in blood pressure when they report being in a negative mood. Low family income and education are related to elevated ambulatory heart rate as is the proportion of blacks in the neighborhood. Paper has been published.
Princeton Ohio School District	Goodman, Adler, McEwen	Study of adolescents from a multi-ethnic school district with a range of SES. The study includes the society and community adolescent versions of the ladders. Analyses show associations between subjective standing and a variety of mental and physical conditions (e.g., perceived stress, depression, obesity). A poster “Developmental trajectories in social status” was presented at Pediatric Academic Societies in 2006 showing that adolescents’ prior year perceived social status and objective SES measures were the major determinants of current year perceived social status, and a curvilinear relationship with income and race by education interaction suggesting that objective SES may be less salient in determining perceived social status among black youth. Completed.
Assessing childhood SES in the Immunization study	Cohen	In the context of this ongoing study of cardiovascular and endocrine reactivity and response to immunization, subjects are being interviewed using the childhood SES, childhood social and physical environment, and adult SES protocols being developed by Sheldon Cohen and the network. The team has administered the childhood SES interview to 180 of an expected 200 subjects. Within 2 months or so, they should have all interviews completed and within 1 year all reactivity and immunization response data collected as well.
Pittsburgh Mind-Body Center: SES and brain activation—childhood and adult SES	Gianaros, Cohen, Matthews	The center supports one R01 type research project on regional brain activation during stress and peripheral physiology (Jennings, Gianaros, and Matthews) and one project supported by pilot funds underway, which examines the association of childhood and adult SES and regional brain activation in response to acute stress (Gianaros, Jennings, Matthews and Cohen). Papers are being reviewed and data will be presented by Gianaros at the May Executive Committee meeting.
Assessing childhood SES in the AHAB study	Cohen	AHAB is a broadly conceived data set including hundreds of psychosocial and biological measures. The network has funded the addition of the childhood SES, childhood social and physical environment, and adult SES protocols (as above), as well as measurements of C reactive protein and IL-6. Interviews have been completed on all 252 subjects. IL-6 production assays have also been completed. Analyses are ongoing with the first emphasis on childhood SES and blood pressure/hypertension medication use in adulthood.
Childhood SES, Cellular Aging and Susceptibility to Infectious Disease	Li-Korotky, Cohen	Study will assess telomere length, and a marker of oxidative stress in a new cold study. A major focus of the study is the role of childhood SES and environments in predicting disease susceptibility. This project will determine whether childhood SES (as well as adult SES, perceived social status and various psychosocial factors) increases infectious disease susceptibility by modulating oxidative stress levels that, in turn, affect the rate of lymphocyte aging and loss of function. The team will do assays as the data come in so should be able to have somewhere between 50-100 subjects completed within a year. Data collection began in April, 2007.
Allostatic load component	Evans, Brody	Gary Evans is collaborating with Gene Brody of University of Georgia to add an allostatic load component to Brody’s longitudinal study of poverty and mental health among rural, African American adolescents in Georgia.
Childhood SES and Disease Trajectories	Cohen	240 subjects will be interviewed about their SES during years 1-18 as well as correlated environmental and behavioral conditions including physical and social characteristics of their neighborhoods, family relationships, and parental health practices. They are then tested in adulthood for sympathetic nervous system and hypothalamic pituitary cortical axis function, immune function, and susceptibility to infectious disease as an adult.
MESA	Diez Roux, Seeman	The Multi-Ethnic Study of Atherosclerosis (MESA) is a medical research study involving more than 6,000 men and women from six communities in the United States. Ana and Teresa were successful in obtaining NIH R01 funding for an add-on to this project to look at the socioeconomic patterning of inflammation and hemostasis using both individual and neighborhood data. AND The network provided funds for a project to assess the relationship between cell\-aging (e.g., telomere length and telomerase) and SES... Due to IRB and lab delays the assays were delayed but the team expects to have the data ready by this summer. Analyses of the data are expected to begin in fall, and manuscripts developed over the coming year.
MacArthur Study of Successful Aging	Seeman, Epel, Blackburn, Cawthon	The team has completed the cell-aging pilot work with a sub-sample of the MacAging population.
Taiwan Aging Study	Seeman	1999 survey on risk factors and disease in elderly Taiwanese. Teresa Seeman is a consultant on the project. The ladders and AL measures are in the battery developed in consultation with the Network. Project completed; a number of papers have been published.
Intervention Studies
Opportunidas(formerly Progresa)	Fernald, Adler	Network supported add-on to large anti-poverty intervention aimed at the poorest 10% of the Mexican population. The team is using the intervention experiment to begin to identify factors mediating the relationship between SES and stress reactivity in children. At this point, intervention communities have received benefits for three years. Fernald presented to the group at our February, 2007 meeting on the relationship between cash transfers and adult obesity. A number of papers have resulted: “Overweight with concurrent stunting in children from rural Mexico: prevalence and associated factors”; “Socio-economic status and body mass index in low-income Mexican adults); “Salivary cortisol levels in children of low-income women with depressive symptoms”); “Predictors of psychological distress among rural Mexican adolescents: A social-ecological analysis); “Correlates of depression in a large sample of Mexican women); “Socio-economic status and blood pressure in a low income population: evidence for a reverse gradient). Three presentations have also resulted: “Innovation in welfare and poverty alleviation: Mexico as a model. Invited address. Poverty, Race and Place”. Berkeley, CA; “Poverty and adult health: Evidence from Mexico”. Invited address. Stanford Institute for Theoretical Economics. Stanford, CA; “Poverty alleviation and child development in Mexico. Invited address”. Office of Behavioral and Social Science Research, National Institutes of Health, 10th Anniversary celebration. Washington D.C.
Opening Doors	Paxson (affiliate member and former core group member), Matthews, Rouse, Lleras-Muney	The project provides community college students with a package of financial assistance, mentoring, and extra help that is designed to increase academic achievement and retention in college. The project has a randomized design. The intervention itself varied across sites, with LA the only site with significant tuition relief. The populations also vary significantly. In NY over half are healthy young men whereas in LA and OH the samples are predominantly women, many of whom are transitioning off of welfare, more of whom were obese. It was a condition of enrollment in LA that the student was a parent. Paxson presented to the group at our April meeting on data related to three of the sites (NY, LA and OH). Around 4,000 have been enrolled (55% treatment and 45% controls). Currently 901 have complete follow-up data. The LA intervention is the only one that appears to be having an effect on educational attainment (e.g., student retention). Within this site, there is evidence of heterogeneity in treatment effects—possibly due to baseline health, but certainly correlated with baseline health (e.g., obesity and smoking). The LA colleges were in New Orleans. 492 participants completed the follow-up interview just before the hurricane. The team is about to go into the field to do a second follow-up survey of this group; they plan to assess whether participation in the program improved coping with Hurricane Katrina and its after-effects.
Experience Corps Trial	Fried, Seeman	A PO1 grant was funded July 2006 through the NIA (Behavioral and Social Research Program) to implement a randomized controlled trial of this program—testing the general hypothesis that participation in Experience Corps will benefit older adults in terms of their social/psychological and physical/cognitive functioning AND that presence of the program in elementary schools will result in benefits to the children with respect to academic achievement and behavioral outcomes. Dr. Seeman is leading the project tasked with assessment of the social/psychological outcomes.
Pathway and outcome studies
Psychosocial Factors, Glucocorticoid Resistance and Susceptibility to Influenza	Cohen	160 healthy adults completed extensive psychosocial batteries and were exposed to influenza virus. Biological markers were collected and susceptibility to influenza monitored to ascertain the effect of psychosocial factors on susceptibility to infection. The glucocorticoid resistance assays were completed last year, allowing the examination of the association of HPA regulation and psychosocial variables, and the role of HPA regulation in disease susceptibility. Initial analyses have not been promising, but further analysis is underway.
Stress reactivity in maternal caregivers of chronically ill children	Epel, Adler	Pilot study to examine AL and cell-aging (e.g., telomere length and telomerase) in a sample under chronic stress, maternal caregivers of chronically ill children compared to controls to examine new and traditional indicators of AL. This landmark study has continued to serve as an impetus for work in this area.
Reserve capacity	Gallo, Matthews	Test of the reserve capacity hypothesis in the context of a project on ambulatory blood pressure of women who are employed in high and low status jobs. Karen Matthews is functioning as a consultant on Linda Gallo’s NIH-funded study examining the influence of SES and ethnicity on women’s cardiovascular risk factors; they are testing the reserve capacity model. Data collection is underway.
Perceived Discrimination, Anger Coping, and Ambulatory Blood Pressure.	Brondolo, Schwartz	This is the first to use ecological momentary assessment techniques to monitor perceived episodes of perceived discrimination in vivo, i.e. as they are occurring in everyday life. The study is designed to test a model of the relationships of perceived racial/ethnic discrimination and ambulatory blood pressure (ABP). The sample will eventually include 500 African-American and 500 Latino adults, 50% female. Participants complete questionnaires to assess trait levels of perceived discrimination, hostility, and anger coping style, and then wear an ambulatory blood pressure monitor and carry an electronic diary that assesses episodes of perceived mistreatment, anger, and anger coping throughout the day. Socioeconomic status (SES) is hypothesized to moderate the relationships of perceived discrimination to impulsive anger-coping and of anger-coping to ABP. Data collection is 75% complete and preliminary analyses are currently being conducted. Presentations were made this year at American Psychosomatic Medicine Society and two manuscripts are under review. Joe is a consultant to this NIH-funded study.
The Relationship of Trait and State Anger to Ambulatory Blood Pressure and Heart Rate Variability	Sloan, Jamner Schwartz	This study is designed to: 1) characterize the influences of situation/context, mood and behavior on autonomic regulation of the heart with an emphasis on social and conflict interactions of adults in everyday settings, and 2) to determine the contribution of individual differences in hostility to these patterns of autonomic cardiac regulation. The goal is to identify distinct patterns of autonomic responsiveness and coronary artery disease (CAD) vulnerability between groups that differ in trait hostility. In addition to trait measures of anger and hostility, ecological momentary assessments of mood, behavior, and situational context are being collected simultaneously with 24-hr ECG and ABP recordings for 120 men and 120 pre-menstrual women, ages 18 to 45, in NYC and Irvine, CA. Data collection is about 65% complete. Joe is a consultant to this 5-year NHLBI-funded study.
Genetics Markers (DNA and mRNA) of Risk for Renal Transplant Rejection and their Ability to Account for Racial Differences	Subthanthiran, August, Schwartz	Researchers are engaged in multiple NIH-funded studies of genetic markers (DNA polymorphisms, mRNA levels of gene expression, and protein levels measured in urine and serum) of acute rejection episodes and renal failure following a renal transplant. African-Americans have much higher rates of acute rejection and renal failure than European-Americans. This research also involves the development of noninvasive tests, based on assessing urinary levels of mRNA, for diagnosing acute rejection that will be cheaper and involve less risk than current method that uses diagnostic biopsies. We have identified several candidate genes and expect to develop risk equations to identify those at greatest risk. Some of this work has been published in 2 NEJM articles. This group has also investigated racial differences in the frequency of genetic polymorphisms for TGF-beta gene expression, and related these to multiple cardiovascular risk factors using data from the Neighborhood Study (see below); one manuscript is under review and another is nearly completed. Joe is the biostatistician for 3 NIH-funded studies.
Masked hypertension: risk factors and consequences project	Schwartz, Pickering	This study focuses on the difference between BP assessed in a clinic and ambulatory BP assessed in people’s everyday environment. The “white coat” effect describes the situation In which BP is substantially higher in the clinic; the “masked hypertension” effect refers to the opposite. In addition to estimating the prevalence rates of these two phenomena, this study seeks to identify predictors of the BP difference between clinic and ambulatory BP, as well as the relationship of this difference to cardiovascular disease risk and target organ damage (e.g., left ventricular mass index). 1000 employed individuals, ages 18-56 will complete a comprehensive battery of psychosocial measures, 3 clinic assessments of BP, a 24-hour ambulatory BP recording, EKG, echocardiogram, and almost all allostatic load measures from the CARDIA ancillary study. Data collection is ongoing. Joe is the PI of this project, one of 3 projects in an NHLBI-funded program project awarded to Dr. Pickering.
The Neighborhood Study	Pickering, Gerber, Schwartz	This study investigates the relationship of SES, neighborhood characteristics (self-reported) to ambulatory blood pressure and sleep quality (in-home polysomnography) in >300 African-American and White NYC residents. Data collection is complete, two conference presentations have been given, and analysis and manuscript preparation are on-going. Joe is a co-investigator of this NHLBI-funded study
Education, Ethnicity and Health	Seeman	Teresa Seeman and collaborators at USC are using NHANES III and IV data to examine ethnicity and time trends in development of health risks, mediated through health behaviors, social support, and psychosocial processes. They are focusing on the ethnic differences in ways education relates to health risks (mediations through behavior and social support). It is anticipated that this work will link up with network analyses of CARDIA data. Several papers have been published; additional manuscripts examining age, gender and ethnic variations in these SES/education gradients have been submitted and others are in preparation.
Pathways Linking Education and Health in Middle Adulthood	Kawachi	The study utilizes the rich information on childhood socioeconomic environment available in the National Collaborative Perinatal birth cohort study to examine the biological mechanisms underlying the relationship between educational attainment and (primarily) cardiovascular disease outcomes in middle adulthood. A variety of potential biological mediators of the relationship between education and health are being measured, including clotting factors, cortisol, lipids, and inflammatory markers.
Educational attainment and Diabetes Control	Adler	Nancy Adler working with collaborators at Kaiser is examining the affect of educational attainment and other socioeconomic factors on diabetic complications in a sample of diabetics in the same health system (so health care coverage is held constant). They are examining different models of how education may affect health and identifying pathways (e.g., differential treatment vs. differential adherence) by which education operates to affect health. Papers resulting from that collaboration include: Schillinger, D., Barton, L.R., Karter, A.J., Wang, F., and Adler, N.E. (2006). Does literacy mediate the relationship between education and health outcomes? A study of low-income population with diabetes. Public Health Reports, 121, 245-254, and Grumbach, K., Braveman, P., Adler, N.E., and Bindman, A.B. (in press). Vulnerable populations and health disparities: An overview. In M. Wheeler, A. Fernandez, D. Schillinger, A. Bindman, Grumbach, T. Villela and T. King. (Eds.), Medical management of vulnerable and underserved patients: Principles, practice and populations. Nancy presented “From bench to bedside to biosphere and back: Diabetes as a case study” at the Clinical and Translational Sciences Workshop: Building a Multidisciplinary and Translational Research Program (CTST), UCSF, San Francisco, CA.
Exploring SES and ethnic barriers to health promoting activities among older adults (Experience Corps sub study)	Luz-Martinez, Fried	The Experience Corps involves putting teams of trained older adults in lower SES elementary schools to serve as mentors. Analyses have shown that the structured social interaction affects educational attainment of the children involved, and the physical, social and cognitive functioning of the long-term low SES seniors. Baltimore’s Experience Corps program has been successful in recruiting and retaining largely African-American women, many who reported incomes on or below the poverty threshold (there is a financial incentive to volunteering); half had not volunteered in the year prior to recruitment. The program has found it more challenging to recruit men of all ethnicities and persons of all SES groups and those who had not volunteered in the past. This qualitative study, using semi-structured interviews in the form of focus groups, collected data on obstacles and potential solutions to participation. Focus groups representing higher and lower SES African American and Caucasian older adults have been completed. Preliminary qualitative analyses indicate that for both men and women social support is at the top of the list as a benefit of participation, as well as getting out of the house and keeping from dwelling on their problems. Barriers to participation are limited funds, inadequate transportation and for the men a disinclination to break their routines. Some of the findings were presented at the 67th Annual Meeting of the Society for Applied Anthropology in March, and a manuscript is currently being drafted (“Ethnic and Class Variations in Health Promoting Activities among Older Adults”) for submission in May. The data were also used for a Master’s CAPSTONE Project focused on the role of health as a perceived barrier and motivator of engagement in social activity. The team is developing an executive summary and some recommendations to share with members of the Commission on Aging and Retirement Education and their affiliates.
Does relative deprivation in income predict health status: a multilevel empirical examination	Subramanian, Kawachi	The primary aim of this study is to examine the independent association between individual measures of relative deprivation and self-rated health in the United States on individual self-rated health. A secondary aim is to test whether individual measures of relative deprivation mediates or modifies the association between state income inequality and individual self-rated health. Analysis used pooled data from the 2003 and 2005 March supplements of the Current Population Surveys conducted by the US Bureau of Labor Statistics. Preliminary results show that irrespective of the variables used to define the reference group, a $5,000 change in relative deprivation (RD) is associated with an increase in the odds of reporting fair-poor health. At the national level the adjusted odds ratio for a $5,000 difference in RD ranges from 1.12 to 1.33 (all at p<.05). At the state level a similar picture is observed with adjusted odds ratios ranging from 1.10 to 1.24. Despite being operationalized as deciles, RD was a significant predictor of reporting fair-poor health with ORs ranging from 1.11 to 1.19. Importantly, the Gini coefficient was a significant predictor of reporting fair-poor health in all fully adjusted models. The OR for a 0.05 unit difference in Gini ranged from 1.17 to 1.19 in the different reference group specifications (defined by various combinations of the variables age, gender, race/ethnicity, educational attainment and state of residence.
Estimating the effects of time-varying income and wealth in the presence of confounded mediators using marginal structural models.	Glymour, Kawachi, Diez-Roux, Williams	Project will treat measures of wealth, income, and employment as time-varying “treatments.” Goal is to estimate the effect of these measures of socioeconomic position on mortality and onset of major health conditions. Because these measures are potentially subject to time-varying confounding, team proposes to use inverse probability weighted marginal structural models (MSMs) to estimate these effects. They will replicate published analyses of the health effects of income using MSMs to estimate the effect of continuous exposures to disadvantaged social position on health using the Health and Retirement Study, a longitudinal data set that has been the subject of influential analyses of the effects of income on health and mortality. The specific aims of this project are to: Estimate the independent effects of income, wealth, and employment on risk of mortality over an 8 year follow-up period (1992-2000) in a sample of middle-aged (born 1931-1941) individuals. Estimate the independent effects of income, wealth, and employment on risk of onset of any major health condition over an 8 year follow-up period. Extend this work, intended to replicate prior studies, to include an additional four years of follow-up now available, include the entire HRS sample born 1947 or earlier, and estimate the direct effect of education on mortality and onset of major health conditions. Work to date has focused on aim 1, estimating the independent effects on mortality from 1992-2004. Because of the very different labor market participation patterns for men and women (and likely different consequences of retirement) we are conducting sex-stratified models. To simplify preliminary models, we have focused on dichotomized low earnings (based on the 25%ile of earnings at baseline, $1800). In crude logistic models the effect of low earnings at wave 2 (1994) on odds of mortality by 2004 (final follow-up) is substantial (OR=2.2, 95% CI: 1.9, 2.7), while the effect of retirement status is more modest (OR=1.25, 1.02, 1.53). The association between low earnings and retirement status, as expected, is very strong (p<0.01), suggesting that ignoring the relationships are likely to induce biased effects. However, the primary effect is from retirement to earnings, thus the change in effect estimates for earnings in preliminary models taking into account the time-varying confounder of retirement, is not tremendously different than the crude effect estimates (HR=2.03, 1.88, 2.17). We are currently running a range of sensitivity tests on this preliminary model and next will jointly estimate the independent effects of earnings and retirement on mortality and each onset of each major health condition.
Socioeconomic Status and Disparities in Periodontal Disease: Mind-Body Pathways	Polk, Cohen	The objective of this proposal is to determine whether psychosocial stress, low trait positive affect, and hypothalamic-pituitary-adrenal axis dysregulation are components of a mind-body pathway through which SES is associated with chronic periodontal disease. They will compare systemic measures of inflammation and the severity and number of active lesions at six months in people low and high in baseline psychosocial stress, trait positive affect, and dysregulated cortisol. They predict that people with baseline psychosocial stress, low trait positive affect, and dysregulated cortisol will have higher levels of inflammatory markers, greater declines in periodontal health, and larger increases in the number of active periodontal lesions six months later. Given the rate of recruiting schedulable participants obtained thus far, the team estimates it will be able to recruit approximately 30 rather than the 50 participants originally proposed. Consulting with network director regarding how to handle this situation. Several available options, including (1) bringing participants back for a third visit 6 months after the second visit; (2) assaying the gingival crevicular fluid and blood for more pro-inflammatory cytokines; (3) easing the eligibility criteria, e.g., including the middle tertile of income or expanding the age range; or (4) returning to the Research Network any unused monies.
Allostatic Load and SES in Costa Rica	Dow	Costa Rican life expectancy is about 78 years, the highest in Latin American (which averages 71 years) and slightly higher than even the United States. This is despite Costa Rica having a per capita GNP one-fifth that of the United States and per capita health expenditure of about one-tenth that of the United States. The country is often regarded as a model for achieving “good health at low cost,” but little is known regarding what population health inputs and behaviors have made this possible, particularly for adults. Specific aims: 1) Supplement a new Costa Rican health survey (CRELES) to include measurement and analysis of C-reactive protein (CRP) as a biomarker for inflammation and stress. 2) Construct an index of allostatic load (including CRP) in the CRELES survey data. 3) Analyze the relationship between allostatic load and socioeconomic, behavioral, and health outcome variables in CRELES. Analysis in progress. Manuscript expected in 2007.
MESA – Salivary cortisol measures: Compliance and reliability	Schwartz, Broderick, Diez Roux, Seeman	An ancillary study of The Multi-Ethnic Study of Atherosclerosis (MESA) collected multiple days of salivary cortisol samples using MEMS caps to assess compliance. The cortisol data collected as part of MESA Stress, an Ancillary Study to MESA, provides a unique opportunity to re-examine issues of protocol compliance, and whether it is necessary/desirable for future studies to use MEMS to monitor compliance, as this dataset is 10 times larger (N=1000) than any previously reported in publications regarding adherence to salivary cortisol sampling. It is also unique in that it includes a high proportion of minority participants, equal numbers of men and women, and a wide socioeconomic distribution. Analysis of this dataset will add substantially to our understanding of participants’ adherence behavior in salivary cortisol collection protocols. We also hope to make recommendations pertaining to salivary cortisol data collection protocols in future research.
Psychosocial stress and mortality	Schwartz	Cochran systematic review (meta-analysis) of all articles dealing with psychosocial stress and mortality. Project began by identifying more than 2000 potentially relevant publications; in the end Schwartz and his graduate students expect o have identified and incorporated approximately 100 relevant publications.
How do subjective indicators of SES relate to objective measures of SES and combine with other psychosocial factors to affect health?	David Williams, James House, Nancy Adler, Ana Diez-Roux	This project will use data from the Chicago Community Adult Health Study (CCAHS), a probability sample of 3,105 adults (aged 18+), sampled from 343 neighborhood clusters (NCs) in the City of Chicago. The study provides a wide range of socioeconomic contexts (low, medium, and high) for all three of Chicago’s largest racial-ethnic groups (983 non-Hispanic whites, 1,240 blacks, and 802 Latinos). The CCAHS has collective reports, observations and archival information from the census and other sources on all 343 NCs and 675 census tracts in which respondents lived. Anthropometric (weight, height, waist, hip, and leg length) and hemodynamic (blood pressure) data were collected from all respondents. It also has biological assays from saliva (cortisol) and blood (C-reactive protein, glycosolated hemoglobin, and total cholesterol) from cooperating respondents in 80 of the 343 NCs. The psychological attributes assessed include a) anger and hostility (anger-in, anger-out, trait anger, cynical hostility and anomie); b) positive and negative affect/beliefs (hopelessness, optimism, pessimism, self-esteem, and depressive symptoms); c) feelings of mastery and control (mastery, dominance, John Henryism). In addition, there are multiple indicators of social relationships, religious participation, chronic and acute stressors, discrimination, early life experiences, and socioeconomic status. Nancy Adler is working with a postdoctoral fellow at Harvard to develop analyses plans and expects to complete the analyses by the end of the year.