Www.publichealth.pitt.edu



Use of a position generator tool to improve national health surveillance surveys of individual social capital (A work in Progress)Jeanette Trauth, Steven Meanley, Todd Baer, Candice Biernesser, Teagen O’Malley and Mudia Uzzi.Abstract: (Submitted to APHA for presentation at November 2016 meeting) Background: Prior evidence suggests socioeconomic status and social support are “fundamental causes” of disease because they embody access to important resources (i.e. money, knowledge, power, prestige and beneficial social connections) that can be used to avoid or minimize negative health outcomes. Given the importance of these critical resources, it’s imperative that we collect data to measure their health impact. Major national health surveys routinely ask questions about an individual’s income, education and occupation but not about an individual’s social capital (ISC). ISC refers to “who you know” and their potential to assist in important problem-solving and decision-making circumstances. The feasibility of collecting ISC information in national health surveys has been demonstrated in other countries by use of “position generators”, which are widely-used research tools. Methods: We reviewed five US national health surveys (BRFSS, General Social Survey, NHIS, NHANES and National Survey on Drug Use and Health) to determine the extent to which questions about ISC are asked. Results: Our review indicates that the majority of social capital measures fail to assess the compositional quality of an individual’s social network. That is, questions aren’t asked about different types of people that an individual knows who possess important resources such as specialized knowledge that can afford them an advantage. Conclusion: We propose that the compositional quality of an individual’s social network can be captured by using position generators. We argue that this is as important to routinely measure at the individual level as income and education to assess social advantage and disadvantage.The following information is required to be in a cover letter for submission to AJPH, which is where we will submit the completed manuscript. ThIs information will also be contained in the body of the manuscript.Our main message: We are making an argument that national health surveillance surveys should routinely ask questions about the nature of an individual’s social network as part of basic demographic information that is collected—similar to level of income and education. Public Health Importance: Social factors such as socioeconomic status and social support are likely “fundamental causes” of disease because they embody access to important resources (i.e. money, knowledge, power, prestige and interpersonal resources embodied in one’s social network), affect multiple disease outcomes through multiple mechanisms, and consequently maintain an association with disease even when intervening mechanisms change.” Major health surveys routinely ask about level of income (but not wealth) and education (which relates to the money and knowledge resources). Surveys also frequently ask about one’s occupation which may relate to the prestige factor. However, the one resource that appears to be missing from the questions asked on large national health surveys is about one’s social connections (i.e. one’s social capital). Contribution to the LiteratureWe are proposing the use of position generator questions, which have not been discussed in the U.S. public health literature, as a way to measure social capital accessed by an individual. “Social capital refers primarily to resources accessed in social networks” (Lin, 1999). These resources are either personal (belonging to an individual) or social (accessed through one’s social network ties). It is through the use of one’s social resources—e. g. a friend’s or acquaintance’s occupational or authority position-- that an individual improves her chances of achieving an advantage of some sort such as finding a job or finding a health specialist. Background We will use the Berkman et al (2000) article as a framework for illustrating how social networks affect health. This will allow us to briefly discuss the various psychosocial mechanisms by which social networks affect health—such as, social support but more importantly—by providing opportunities to access resources (e.g. information) embedded in the social network. Then transition to the following.Social CapitalSocial and behavioral scientists have widely recognized the roles social relationships play in influencing health behaviors and outcomes. Social capital has received attention within public health contexts given its capacity to measure the health-promoting and detrimental characteristics of an individual’s social network (Domínguez & Arford, 2010). It is commonly characterized as an interpersonal construct focused on cooperation, trust, and reciprocity, operating with the intention of creating shared benefits (e.g., access to resources; Veenstra et al., 2005). Colloquially-speaking, social capital is often referred to as the “who you know” quality of a social network, especially with regard to how it relates to accessing information and advice to assist an individual problem-solving or making important decisions (Luthans et al., 2004). Social capital exists across three dimensions – bonding (strength of existing relationships), bridging (building new relationships), and leveraging capital (fostering linkages between community members and community organizations; Walter & Hyde, 2004). (NOTE: Need to address conceptualizations of the levels of social capital: group vs individual. Colman, Putman, Kawachi, Bourdieu. See Borgatti et al., 1998.)Vertical Access of Social ResourcesIn accordance with our fundamental cause hypothesis, we focus on leveraging social capital. Leverage-producing connections typically concern the vertical access of social resources, specifically resources like information and opportunities for advancement that are normally unavailable to the individual if the connection did not exist (Domínguez & Arford, 2010). Social Capital through Compositional QualityWe argue that understanding the capacity for individuals and communities to engage in upward mobilization as it relates to accessing social resources is critical for developing behavioral health interventions at multiple levels. Largely understudied in the social capital literature is the concept called compositional quality. Compositional quality refers to the number of within-group members who possess certain qualities, such as high intelligence, wealth, or power (Borgatti et al., 1998). This construct builds upon the benefits of having heterogeneity within one’s social network, prioritizing the presence of specific characteristics deemed contextually important (Lakon et al., 2008). In terms of social relationships and health, an account of compositional quality may elucidate the social connectedness of individuals and communities to individuals who hold positions of power, prestige, expertise, and wealth in the health sector. We suggest that having access to these resources potentiates improved quality of problem-solving and decision-making in navigating complex health problems and systems.Measuring Compositional Quality through Position GeneratorsOne potentially feasible method to account for compositional quality in a public health context is through adapting the use of position generators (Lin & Dumin, 1986). Position generators have been largely implemented to determine the number of contacts an individual has specific to structural positions salient in a society (Lin, 1999). Segue to next section below.Whereas some theorists conceptualized social capital as benefits afforded at the group, community, or societal level, Lin’s view of social capital is that of an individual-level attribute and indicative of the resources embedded in one’s social structure that are “accessed and/or mobilized in purposive actions” (Lin, 2001). Compared to household and community level contexts of social capital, measures of individual level social capital have been shown to have the greatest influence on health (Giordano, 2011).In recent years, public health has used network analysis tools to better understand the sociostructural aspects of health (Luke,2007; Moore 2009, 2011, Carpiano 2011). Position generators can be used to measure the compositional quality of an individual’s social network and, by extension, quantify social capital at an individual level to better understand how network resources (potential or actualized) influence health and health behavior.Insert description of Position Generators-- Lin (2001)--measurement techniques for investigations of social capital.Psychometrics of Position Generators (PG). Parallel-Forms Reliability: Verhaeghe (2012) tested the reliability of PG using two different lists of occupations and found that of 13 PG measures tested only the volume measure, which is the total count of accessed occupations, had good reliability. Other measures tested compositional quality of networks including measures of occupational prestige and SES, and measures of social class. These measures had poor and fair to good reliability respectively. Limitations: sample of students. Parallel tests. Test-Retest Reliability: see Angelusz and Tardos, Erickson 2004 Construct Validity: Moore (2009) used data from the Montreal Neighborhood Survey of Lifestyle and Health that included position generators to measure individual level social capital by mapping an individual’s social ties to specific occupations to examine the association of social capital to abdominal adiposity and BMI. A total of 16 occupations were selected for the position generator and each participant was asked if they knew someone by first name who would fit into each of the 16 occupations. Each of the 16 occupations was attached to a prestige score which allowed for the calculation of potential resources available to the participants via his/her social ties. Three social capital indices were derived from the positions generator measure and included upper reachability, range, and diversity of a participant’s social network. All three indices were found to be associated to waist circumference and body mass index such that those with greater reachability, range and diversity were less likely to have at-risk WC and BMI. Yang (2013) used position generators to study the relationship between social capital and health literacy in a probability sample of 1,100 Taiwanese residents and found that social capital, defined in terms of range, extensity, and upper reachability, is a strong predictor of health literacy. Of note, income was not significantly associated with health literacy. This underscores the importance of measuring multiple social determinants of health! In convergent/divergent validity testing of community belongingness, Carpiano and Hystad (2011) found that social network diversity measures derived from position generators were associated with a sense of community belongingness and measures of self-related health. Those that reported greater diversity in their network also reported a greater sense of community belongingness and overall better self-rated health. Additional information to add: Discuss the development and use of the position generator broadly, but more specifically in the context of public health and surveillance. This section will also discuss the psychometric properties of the instrument, challenges to using the method, and feasibility of inclusion of this method in population based surveys. Methods To address the need for inclusion of social capital measures in public health research we examine the prevalence of social capital and more broadly social support measures in major population-based surveillance surveys conducted in the U.S., Canada, Britain, and Australia. Major World Health Organization surveys are also included to provide a more global perspective. The purpose of this inventory is to account for what information we have and what information is needed to advance our understanding of how social capital, as a fundamental determinant of health, influences population health and behavior, healthcare access, and patterns of health disparities in the U.S. and abroad. Explain the rationale for the surveys we selected to review. We will review national behavioral surveys—e.g. BRFSS—in the US, Canada and other countries—approximately 10 surveys in total.We are looking for questionnaire items that address the social network constructs discussed in our background above. ResultsPresent the results of what we found in our review of national surveys. This can be done succinctly probably in a series of tables showing an inventory of what various surveys are measuring. See the following table as an example.An initial review of 5 national surveys reveals that the majority of social capital measures fail to assess the compositional quality or imbedded resources of an individual’s social network.SurveySocial Capital MeasureStructural Level (Individual vs Collective)Cognitive Level (Interpersonal Trust & Reciprocity)BRFSSSocial SupportIndividualYesGeneral Social SurveySocial NetworkIndividualNoNHISSocial NetworkIndividualNoNHANESSocial NetworkIndividualNoNational Survey on Drug Use and HealthNeighborhood Social SupportSocial NetworkCollectiveIndividualYesDiscussion The following are points for inclusion in the Discussion section –in no particular order.?Discuss the utility of collecting information about an individual’s social network for intervention purposes. Provide examples of how researchers/ interventionists have altered people’s social networks to improve health outcomes—either directly—e.g. cancer support groups, or indirectly– by lifting people out of poverty—e.g. the national Circles USA initiative. ?Discuss the implications of collecting this type of information in terms of a life course perspective and the elimination of health disparities. The Life course Perspective offers insight into the value of measuring social advantage by determining key time periods where increased knowledge of social condition could define priorities for prevention and intervention. This approach hypothesizes that there is a social trajectory of disease, when there are key moments of vulnerability that, if altered, could decrease long-term susceptibility to poor health. This perspective has been thoroughly studied in chronic disease epidemiology showing linkages between exposure to poor social conditions and chronic conditions which contribute to mortality across all stages of the lifespan from gestation to late life (Ben-Schlomo, et al, 2002; De Genna, et al, 2006; and Lynch, et al 2005).There are many such turning points present within one's life that have the potential to impact long-term health status, including having a first child, getting married/divorced, and the transition to retirement.A study conducted by De Genna and colleagues (2006) illustrates the potential for salient time periods in one's life to be targeted for prevention/intervention activities by providing evidence of inter-generational impacts of social disadvantage. A cohort of mothers was studied who, during their childhood, experienced social isolation, poverty, and poor access to healthcare. This turbulent upbringing led to an adolescence fraught with risk taking behavior, such as smoking, drinking, violence/aggression, drug use, and becoming a teenage mother co-occurring with lower educational attainment and patterns of social withdrawal. Strikingly, the impacts of early social disadvantage on these mothers left an impact on the health of their children in utero, being more likely to experience problems during pregnancy (OR=1.44; p<.08), and during the early stages of their children's lives, most markedly through a predictive relationship between maternal smoking and diagnoses of respiratory illnesses in their children (OR=1.07; p<.01). This cross-generational evidence of the ramifications of poor social condition, further exemplifies the need to carefully measure critical periods within the life course.In addition to the life course, policy makers and clinicians may utilize this measurement tool to prioritize intervention with those most vulnerable to health disparities. The need for attention to health disparities is clear in a climate where minorities, and in particular African Americans, have higher rates of obesity, diabetes, hypertension, heart disease, and higher incidence and mortality rates from many cancers that are amenable to early diagnosis and treatment. The linkage to this disease burden co-occurs with increased risk of social disadvantage, where African Americans, Hispanics, and other minorities are less likely to receive adequate education or access to healthcare and more likely to be in poverty and stigmatized or discriminated against (CDC, MMWR). Minorities are very clearly more likely to have disadvantageous social environments than their Caucasian counterparts, and correspondingly are more vulnerable to risk of poor health. This measure will allow for the screening of social condition on a mass level to identify needed areas of change for those who are at greatest need. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download