While debated , components of social capital include reciprocity, trust, safety, social agency, social networks, value of life, and employment connections . Research on social capital and HIV has predominantly focused on preventing HIV transmission with increased social capital being associated with decreased HIV transmission . High social capital was the strongest predictor of HIV self-management in WLHIV and specifically on HIV medication adherence . Social capital has also been negatively associated with substance use among youth and African-Americans . Discrimination was associated with illicit drug use in African American women and aspects of social capital protected women against its effects . However, in rural Appalachia illicit drug use was associated with greater social capital , highlighting the need to carefully assess social capital and its relationship to substance use behaviors. Given that most examinations of social capital among PLHIV are quantitative and cross-sectional, there is a need to gather more qualitative and mixed methods data to better understand its relationship to health outcomes . We conducted a 6-month, mixed methods longitudinal study with 29 WLHIV who were either current or previous illicit substance users. The use of a concurrent mixed methods longitudional design was necessary because while numerous studies document a relationship between social capital and HIV health outcomes , little is known about how the dimensions of social capital are developed and used to improve those outcomes over time. Integrating concurrent qualitative and quantitative data to examine social capital in this population will allow a more in-depth understanding of how social capital influences HIV self-management and substance use patterns among WLHIV.
Social capital cannot be fully captured by psychometric scales,vertical growing system but by integrating qualitative data with quantitive measures we can more comprehensively describe its dimensions. Potential subjects were from an existing HIV-research registry of approximately 300 adults living with HIV in Northeast Ohio. All registry participants had contacted the study team about previous research opportunities and had given written consent to be included in the registry. All women in the registry were sent an IRB-approved letter explaining the study and asking them to contact the study team via telephone if they were interested in participating. Those who responded were screened via telephone for illicit drug use using the Drug Abuse Screening Test-10 . Subjects scoring > 1 were included in the current illicit drug use strata and those scoring a 0 were included in the not using illicit drugs strata. If we found out during the qualitative interview that the participant has used illicit substances within the past 12 months, she was given a DAST score > 1 regardless of the screening form or the surveys. If a woman met eligibility criteria, she was scheduled for a research visit. At the first research visit, a research assistant explained the study and obtained written informed consent. Next, she completed a battery of surveys in REDCap and an open-ended social capital interview. Interviews were guided by a semi-structured interview guide and audio-recorded. Quantitative surveys were completed prior to interviews to introduce the concepts that were to be discussed and to standardize the data collection. At the conclusion of the visit, participants were compensated with $20 cash for their time and travel. Data were collected between July, 2015 and June, 2016. All procedures were approved by the Institutional Review Board at the Medical Center.
To quantitatively describe how social capital influences HIV self-management and substance use patterns over time among WLHIV we assessed the following variables, based on the literature described above: social capital, HIV self-management, substance use, and potential confounding variables. Participants completed study assessments at baseline and then approximately three and six months later. Social capital was measured using the 36-item Social Capital Scale. This widely-used and psychometrically validated instrument generates a total social capital score and measures eight sub-scales including: participation in the local community, social agency, feelings of trust and safety, neighborhood connections, friends and family connections, tolerance of diversity, value of life, and workplace connections . Participants rated each item on a 1–4 Likert-type scale. Higher mean scores indicate more social capital. Cronbach’s alpha reliability for the social capital scale in adults living with HIV is 0.88 . We examined two aspects of HIV self-management, HIV medication adherence and a global measure of HIV self-management. HIV antiretroviral medication adherence was assessed with a 30-day adherence visual analog scale . To measure HIV self-management more globally, participants also completed the 20- item HIV Self-Management Scale, which generates a total score from items measured on a 0–3 scale and measures three domains of HIV self-management . This scale has been previously examined and was found to be psychometrically valid for use among WLHIV . Substance use was assessed with the valid and reliable 11-item self-report Drug Use Disorders Identification Testdeveloped to screen individuals for drug problems . Total scores range from 0 to 44, with higher scores being suggestive of a more severe drug use problem.
Descriptive and potential confounding variables included demographic and medical characteristics, traumatic events and experiences of discrimination and were selected based on previous literature reviewed above. Demographic characteristics were self-reported and included, race, education level, family composition, employment, sexuality and housing status. Medical characteristics, abstracted from the participant’s electronic medical record, included year diagnosed with HIV, current CD4+ T cell count, HIV Viral Load, HIV medication history, and retention in HIV primary care. Recent traumatic events were assessed with the 20-item List of Threatening Experiences Scale, which lists the experience of traumatic events in the past month. Each of these 20-items are summed and higher scores indicate more traumatic experiences . Experiences with discrimination were assessed with the widely-used, and valid 9-item Everyday Discrimination Scale . Participants noted how often they experienced acts of discrimination. All items are summed and higher scores indicate more discrimination. To qualitatively describe how social capital influenced HIV self-management and substance use patterns, we developed a semi-structured qualitative interview guide based on existing literature to guide in-person interviews. Prior to using it with participants, experts in substance use and women and in WLHIV assessed the interview guide for clarity, relevance, and appropriateness. The baseline interview guide initially focused on early substance use, current health practices, past and current social networks, and social capital dimensions . After several interviews, based on new themes spontaneously emerging from the interviews, it was revised to include more probes related to how trust influences health behaviors and the specific role of faith and its influence on health behaviors. In other words,plant growing rack in the first few interviews WLHIV discussed the role of trust and faith on their health behaviors and we deemed it so important that we wanted to give all respondents a chance to discuss these topics. The threeand six-month interview guides were shorter and focused on changes in life situations, health practices, social capital, substance use and resilience that were observed in their quantitative measures. These guides were designed to help us understand any changes in our variables of interest and how they influenced self-management behaviors. Data analysis of the qualitative and quantitative data occurred at the same time, but were not integrated until both types of data were analyzed. In analyzing the quantitative data, we first assessed the distribution of all quantitative variables . We summarized baseline characteristics by using means, standard deviations, medians, interquartile ranges, counts and percentages of women by substance use group , depending on the variable’s distribution. We used generalized estimating equations with an identity link function and an unstructured correlation structure to describe how social capital and substance use influences HIV self-management across the three time points. Separate models were fit for each HIV self-management outcome. In addition to the effect of social capital and substance use, we examined independent effects of age, discrimination and traumatic events by adding these covariates to GEE models.
All statistical analyses were conducted using Stata 14.0 with p-values < 0.05 considered statistically significant. Qualitative data were managed using the qualitative data analysis program Dedoose and was analyzed by the research team using qualitative description methodology . Data were transcribed and examined by two research team members who coded the data using the constant comparative method, identifying patterns and themes . These team members met regularly during coding to discuss consistencies and inconsistencies in the data. A priori codes related to social capital, substance use, and self-management based on our literature review, were initially applied then inductive codes were applied. Transcripts were revisited in a series of iterative steps to confirm coding classification and that theoretical saturation was reached. Variations on the themes and negative cases were identified to help understand the full range of data within codes. A final code book of themes, definitions, and exemplar codes was created to aid analysis. Data were coded, and analyzed using Dedoose version 8.0.42 . Study procedures are presented consistent with the Good Reporting of a Mixed Methods Study standards .State governments, in particular, play prominent roles governing critical sectors of the economy like energy, healthcare, and education. This, along with the dysfunction, polarization, and gridlock that has plagued federal politics in recent years, has made state governments attractive targets for reformers. Polarization and gridlock, of course, extend to the sub-national level, but the sheer variety in the political landscapes across the states means there are more opportunities to shift public policy. Some states also have institutions like the ballot initiative that make it easier bypass legislatures and break the status quo. While federalism and “states’ rights” has traditionally been a clarion call for conservatives seeking to preserve the status quo, there is now a movement on the Left to leverage the states to drive forward progressive reforms . Yet, relying too heavily on the states presents its own pitfalls. State policy can be an impotent tool for addressing complex political economy issues of broad scope like climate change. For one, taking on these issues often requires coordination across the federal system to address spillovers—meaning federal action is critical. Second, states also tend to be budget constrained, which restricts their ability to enact major policy change. They, unlike the federal government, do not control their own currencies, and are required to, for the most part, run balanced budgets. Attempts to raise more revenue through higher taxes risks chasing businesses and wealthier residents to lower-tax jurisdictions. Third, and finally, reduced legislative professionalism and a lower density of countervailing interests means that state politics can be a more favorable venue than national politics for incumbent economic interests—which often work to block reforms that challenge their position. The potential for policy reform in the states, combined with the limitations of state policy for addressing major issues, motivates examining political and policy inter dependencies across the federal system. A reform achieved in one state alone might make just a small dent in a problem of broad scope, but if it leads other states and the federal government to follow suit, the overall effects could be significant. Thus, this dissertation addresses the following general question: how does adoption of policy reforms in the states affect the political prospects for those reforms in the broader federal system? It is, of course, not the first to examine questions of policy interdependence in American federalism. The innovation is in the theoretical and empirical approaches, and the particular focus on political economy—interactions between elements of the political and economic systems. The conventional framework for studying policy interdependence in American federalism is policy diffusion, which refers generally to how adoption of a policy in one unit increases the likelihood of its subsequent adoption elsewhere . This general idea can be traced at least as far back as Louis Brandeis, who famously, in a 1932 judicial opinion, suggested that states could function as “laboratories of democracy” —state experimentation could be leveraged to identify promising policy designs, which might then be propagated. As the policy diffusion literature has developed, scholars have productively moved from simply tracing the spread of policies to trying to identify the mechanisms driving that spread. Two mechanisms have emerged in the federalism literature as central to the diffusion process. The first, which Brandeis alluded to, is policy learning . Because lawmakers generally aim to pursue policies that work well for constituents, they might, in considering which policies to adopt, choose policies that have already been shown to work well in other units .