Contrasting speed and accuracy approaches to measure executive functions in three low-and middle-income countries.

Executive functions (EF) can be measured by tests assessing accuracy, reaction times and by computing scores which combine these two components. Interpretation issues can arise from the use of different scoring methods across studies. Given that EF measures and their scoring methods are predominantly developed and validated in high income countries, little is known about the generalisability of such methods cross- culturally. The current paper compares two different established scoring approaches for measures of inhibition and cognitive flexibility: difference scores (which utilise reaction time only) and computed scores (combining accuracy and reaction time). We utilised data collected in adulthood from three low- and middle-income birth cohorts (Guatemala, Philippines, South Africa). Non-normal distributions were observed for both scoring methods in all three samples; however, this was more pronounced for the difference score method. Differing distribution patterns were observed across the three cohorts, which was especially evident in the Guatemala cohort, highlighting potential issues with using these methods across diverse populations. The data suggest that the computed scores may be a reliable measure of EF. However, the different ways of scoring and interpreting EF instruments need to be considered carefully for each population before use.

Development and validation of the Soweto Coping Scale: A mixed-methods, population-based study of adults living in Soweto, South Africa.

Mental health disorders are amongst the leading contributors to the burden of disease and need to be prioritised in policy making and program implementation. In the absence of mental healthcare, people often navigate their own social support and activate individual coping mechanisms to sustain their emotional well-being. Few South African studies conceptualise and evaluate the strategies people use to manage adverse situations in non-clinical samples. We conducted two related ethnographic studies of stress and coping in Soweto (n = 107). We then used the studies to develop a novel scale to measure local forms of coping and evaluated its use in an epidemiological surveillance study (n = 933). In a split sample analysis, we first conducted exploratory factor analyses and then a comparative fit index assessment. In the exploratory factor analysis, we obtained a two-factor solution: problem-focused/emotional coping and religious coping. In the confirmatory factor analysis, both domains had good model fit above the conservative ≥ 0.95 cut-off, and both factors had adequate internal consistency (religious coping = 0.72; problem/emotion focused coping = 0.69). Both the problem-focused/emotional and the religious coping subscales were positively correlated with quality of life, except that the religious coping subscale was not correlated with social relationships. Total adverse childhood experiences were correlated with the problem-focused/emotional coping subscale but not with the religious coping subscale. We conclude that the Soweto Coping Scale provides a novel understanding of local forms of coping and can be used by mental healthcare researchers and providers who seek to develop interventions for promoting mental health and social well-being.

A mixed-methods, population-based study of a syndemic in Soweto, South Africa.

A syndemic has been theorized as a cluster of epidemics driven by harmful social and structural conditions wherein the interactions between the constitutive epidemics drive excess morbidity and mortality. We conducted a mixed-methods study to investigate a syndemic in Soweto, South Africa, consisting of a population-based quantitative survey (N = 783) and in-depth, qualitative interviews (N = 88). We used ethnographic methods to design a locally relevant measure of stress. Here we show that multimorbidity and stress interacted with each other to reduce quality of life. The paired qualitative analysis further explored how the quality-of-life impacts of multimorbidity were conditioned by study participants' illness experiences. Together, these findings underscore the importance of recognizing the social and structural drivers of stress and how they affect the experience of chronic illness and well-being.

Development and validation of the Soweto Stress Scale: Mixed-methods, population-based study of adults living in Soweto, South Africa.

This article moves from ethnography to epidemiology to generate a locally specific assessment of stress and evaluates it correlates with general psychiatric morbidity. We conducted two related ethnographic studies of stress and distress in Soweto, South Africa (N=107). We then used these studies to develop the Soweto Stress Scale, piloted the scale, and then evaluated its use in an epidemiological surveillance study of stress and health (N=933). We used factor analyses to evaluate factor structure and maximum likelihood estimates to evaluate comparative fit indices. The Keiser-Meyer-Olkin test identified sufficient sample variation, and the scale was suitable for factor analysis. The confirmatory factor analysis supported a single-factor model with a χ2(df) (104) = 475.88; p < 0.001. Even though the comparative fit index/Tucker-Lewis index were poor and could not be improved, the Cronbach's alpha (0.81) was excellent, suggesting that the model was a reasonable representation of the data. The final model indicated that there was covariance between items and consistency with our theoretical concept of stress. The Soweto Stress Scale shows strong internal consistency and reflects a very localized view of social stressors and may serve to identify those with higher psychological morbidity. Given the racial, ethnic, and linguistic diversity within South Africa, our emic stress scale was tested in a community sample but may be useful for screening individuals with higher levels of stress or psychological distress within clinical settings.

Stunting in infancy, pubertal trajectories and adult body composition: the Birth to Twenty Plus cohort, South Africa.

BACKGROUND/OBJECTIVES: Childhood rapid growth and earlier puberty onset have been associated with adult obesity. However, the association between childhood stunting, pubertal timing and adult obesity is unclear. We examined whether the relationship between stunting at age 2 years (y) and body composition at 23 years is mediated by adolescent body mass index, and pubertal development, using the Birth-to-Twenty Plus cohort (South Africa). SUBJECTS/METHODS: For 1036 participants, data on anthropometrics between birth and 23 years, maternal factors, and pubertal development (Tanner scale at 9-16 years) were collected. Stunting at 2 years (height-for-age z-score < -2), 5-18 years BMI-for-age trajectories, pubertal development trajectories, and DXA-derived fat mass (FM) and fat free mass (FFM) at 23 years were determined. Data were analysed using hierarchical regressions and structural equation models. RESULTS: Stunting was directly associated with slower pubertal development and with shorter adult stature, but was not associated with adolescent BMI trajectories, adult FM or FFM. However, stunting was indirectly associated with adult FM and FFM through the direct associations between slower pubertal development and lower FM and between shorter height and lower FFM. BMI trajectories were independently associated with FM and FFM. CONCLUSIONS: Being stunted in this population predicted adult body composition through slower pubertal development and shorter adult stature.

Executive functions form a single construct and are associated with schooling: Evidence from three low- and middle- income countries.

Measuring executive function (EF) among adults is important, as the cognitive processes involved in EF are critical to academic achievement, job success and mental health. Current evidence on measurement and structure of EF largely come from Western, Educated, Industrialized, Rich and Democratic (WEIRD) countries. However, measuring EF in low-and-middle-income countries (LMICs) is challenging, because of the dearth of EF measures validated across LMICs, particularly measures that do not require extensive training, expensive equipment, or professional administration. This paper uses data from three LMIC cohorts to test the feasibility, validity and reliability of EF assessment in adults using three sub-tests (representing key components of EF) of the NIH Toolbox Cognitive battery. For each cohort, all three EF measures (inhibition, flexibility and working memory) loaded well onto a unidimensional latent factor of EF. Factor scores related well to measures of fluid intelligence, processing speed and schooling. All measures showed good test-retest reliability across countries. This study provides evidence for a set of sound measures of EF that could be used across different cultural, language and socio-economic backgrounds in future LMIC research. Furthermore, our findings extend conclusions on the structure of EF beyond those drawn from WEIRD countries.

Intimate Partner Violence and Its Association With Self-Determination Needs and Gender-Power Constructs Among Rural South African Women.

This study aimed to identify psychosocial correlates of intimate partner violence (IPV) by using constructs derived from the self-determination theory (SDT) and gender-power scales. Cross-sectional data (N = 238) were collected from women in the Eastern Cape, South Africa, and were used to test a structural equation model (SEM). The majority (87%) of the participants reported having sexual partners in the past 3 months, and in terms of IPV victimization, 36% and 26% of women had ever experienced verbal and physical abuse, respectively. Bivariate correlations showed that autonomy and beliefs about gender equality (BGE) were strongly associated with IPV. This finding was also confirmed in the SEM analysis, which indicated that autonomy had a direct effect on IPV suggesting that women who are in relationships that allow them to make decisions along with their partners possibly experience less IPV. In addition BGE, which was hypothesized to play a mediating role showed a significant direct association with IPV, suggesting that women who are aware of their rights may experience less IPV by choosing partners who do not espouse hegemonic masculinities or strong patriarchal beliefs. Our findings suggest that it would be important to incorporate decision-making skills and human rights awareness in future community-based sexual health and reproductive rights interventions.

Understanding the role played by parents, culture and the school curriculum in socializing young women on sexual health issues in rural South African communities.

BACKGROUND: the decline in South Africa's HIV infection rates especially among young women is encouraging. However, studies show that the 15-24-year-old cohort remains vulnerable. As they still report early sexual debut, being involved in sexual partnerships with older men as well as having unprotected sex. These risky sexual behaviors may be linked to factors such as the parent-child sexual health communication and the timing of the first talk. The quality of sexual health information received in school may also be important for enhancing healthier sexual behaviors. AIMS AND OBJECTIVES: to investigate the what, when and how sexual health communication occurs in rural South African families and to determine whether such communication patterns have changed over time. We also wanted to get an in-depth understanding of the roles played by culture, sexual health education and peers in the socialization of young women on sexual matters. METHODS: a purposive sample of (n = 55) women who were 18-35 years old was selected and interviewed in focus group discussions (FGDs). RESULTS: the FGD findings show that parent-child communication on sexual matters in rural communities is limited to messages that warn against pregnancy. It is also laden with cultural idioms that are not well explained. The school sexual health curriculum also fails to adequately equip adolescents to make informed decisions regarding sexual matters. All this seems to leave room for reception of misguided information from peers. CONCLUSIONS: findings highlight a need for designing interventions that can create awareness for parents on the current developmental needs and sexual behavior of adolescents. For adolescents programs would need to focus on providing skills on personal responsibility, and how to change behavior to enhance sexual health.

Self-determination and gender-power relations as predictors of condom use self-efficacy among South African women.

This article identifies correlates of condom use self-efficacy using concepts from self-determination theory and gender-power measures. A cross-section of Xhosa-speaking women (n = 238) from Eastern Cape, South Africa, was used to conduct bivariate correlations and multivariate linear regression analyses. Gender equality beliefs and HIV knowledge were positively associated with condom use self-efficacy generally and in risky situations. Condom use self-efficacy generally was also positively associated with power balance attitudes, negative beliefs about intimate partner violence, and positive growth perspective, while the association with hopeless personal perspective was negative. Surprisingly, lack of social support was positively associated with condom use self-efficacy in risky situations. The predictors of condom use self-efficacy identified in this study that may serve as change objectives for future sexual health promotion interventions.