Search

Network-based strategies and competencies are probably even more important in poor societies with limited means of communication and less effective formal structures than in developed economies. And they often deal with life and death matters. This paper presents lessons from and insights about the nature of and the impacts of informal social networks in reducing fertility and coping with HIV/AIDS in Kenya and Malawi based on analyses of quantitative longitudinal data and qualitative data that the authors and their collaborators have been collecting and analyzing for over a decade. Specific lessons include the relevance of social networks and informal interactions for many different domains related to health – and thus life and death – in developing countries, the importance of accounting for the endogeneity of network partners in analyzing network effects, that networks are important even with control for endogeneity, that network effects may be nonlinear, that there may be multiple equilibria, that which networks may either reinforce the status quo or help diffuse new options and behaviors, that both the context (e.g., the degree of market development) and the density of networks matter (possibly interactively), and that multiple approaches, including both qualitative and quantitative analyses, can be informative in providing more in-depth understanding of what networks do and how they function.

Objective: To examine the acceptance of repeat population-based voluntary counselingand testing (VCT) for HIV in rural Malawi.Methods: Behavioral and biomarker data were collected in 2004 and 2006 fromapproximately 3,000 adult respondents. In 2004, oral swab specimens were collected andanalyzed using enzyme-linked immunosorbent assay (ELISA) and confirmatory Westernblot tests while finger-prick rapid testing was done in 2006. We use cross-tabulationswith chi-square tests and significance tests of proportions to determine the statisticalsignificance of differences in acceptance of VCT by year, individual characteristics andHIV risk.Results: First, over 90% of respondents in each round accepted HIV test, despitevariations in testing protocols. Second, the percentage of individuals who obtained theirtest results significantly increased from 67% in 2004 when the results were provided inrandomly selected locations several weeks after the specimens were collected, to 98% in2006 when they were made available immediately within the home. Third, whereas therewere significant variations in the socio-demographic and behavioral profiles of those whowere successfully contacted for a second HIV test, this was not the case for those whoaccepted repeat VCT. This suggests that variations in the success of repeat testing mightcome from contacting the individuals rather than from accepting the test or knowing theresults.Conclusions: Repeat HIV testing at home by trained health care workers from outsidethe local area, and with either saliva or blood, is almost universally acceptable in ruralMalawi, and thus likely to be acceptable in similar contexts.

Network-based strategies and competencies are probably even more important in poor societies with limited means of communication and less effective formal structures than in developed economies. And they often deal with life and death matters. This paper presents lessons from and insights about the nature of and the impacts of informal social networks in reducing fertility and coping with HIV/AIDS in Kenya and Malawi based on analyses of quantitative longitudinal data and qualitative data that the authors and their collaborators have been collecting and analyzing for over a decade. Specific lessons include the relevance of social networks and informal interactions for many different domains related to health – and thus life and death – in developing countries, the importance of accounting for the endogeneity of network partners in analyzing network effects, that networks are important even with control for endogeneity, that network effects may be nonlinear, that there may be multiple equilibria, that which networks may either reinforce the status quo or help diffuse new options and behaviors, that both the context (e.g., the degree of market development) and the density of networks matter (possibly interactively), and that multiple approaches, including both qualitative and quantitative analyses, can be informative in providing more in-depth understanding of what networks do and how they function.

Objectives: Earlier studies suggest that being married in later life protects against dementia, and that being single in old age increases the risk of dementia. In this study, we examine midlife marital status trajectories and their association with dementia and mild cognitive impairment (MCI) at ages 70 plus using a large population based sample from Norway. Methods: Based on a general population sample linked to population registries (N = 8706), we used multinomial logistic regression to examine the associations between six types of marital trajectories (unmarried, continuously divorced, intermittently divorced, widowed, continuously married, intermittently married) between age 44 and 68 years from national registries and a clinical dementia or a MCI diagnosis after age 70. We estimated relative risk ratios (RRR) and used mediation analyses adjusting for education, number of children, smoking, hypertension, obesity, physical inactivity, diabetes, mental distress, and having no close friends in midlife. Inverse probability weighting and multiple imputations were applied. The population attributable fraction was estimated to assess the potential reduction in dementia cases due to marital histories. Results: Overall, 11.6% of the participants were diagnosed with dementia and 35.3% with MCI. Dementia prevalence was lowest among the continuously married (11.2%). Adjusting for confounders, the risk of dementia was higher for the unmarried (RRR = 1.73; 95% CI: 1.24, 2.40), continuously divorced (RRR = 1.66; 95% CI: 1.14, 2.43), and intermittently divorced (RRR = 1.50; 95% CI: 1.09, 2.06) compared to the continuously married. In general, marital trajectory was less associated with MCI than with dementia. In the counterfactual scenario, where all participants had the same risk of receiving a dementia diagnosis as the continuously married group, there would be 6.0% fewer dementia cases. Discussion: Our data confirm that staying married in midlife is associated with a lower risk of dementia and that divorced people account for a substantial share of dementia cases. [ABSTRACT FROM AUTHOR]