Non-response in a national health survey in Germany: An intersectionality-informed multilevel analysis of individual heterogeneity and discriminatory accuracy

What does the MAIHDA method explain?

Multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) is a new approach to quantitative intersectional modelling. Along with an outcome of interest, MAIHDA entails the use of two sets of independent variables. These include group demographics such as race, gender, and poverty status as well as strata which are constructs such as Black female poor, Black female wealthy, and White female poor. These constructs represent the combination of the demographic variables. To operationalize the approach, an initial random intercepts model with strata as a level 2 context is specified. Then, another model is specified that includes the strata as well as the demographic variables as level 1 fixed effects. As such, it is argued that MAIHDA uniquely identifies the additive and intersectional effects for any given outcome. In this paper we show that MAIHDA falls short of this promise: the strata are an individual-level composite variable not a level 2 context. Rather than being analogous to neighborhoods as contexts, strata are analogous to socio-economic status which is a combination of individual-level demographic variables, albeit often presented as a group-level characteristic. The result is that the demographic variables are inserted in both level 2 and 1. This duplication across the levels in MAIHDA means that there is a built-in collinearity across the levels and that the models are mis-specified and, therefore, redundant. We conclude that single-level models with the demographic variables and interactions or with the strata as fixed effects are still the more accurate models for quantitative intersectional analyses.

Health inequalities at the intersection of multiple social determinants among under five children residing Nairobi urban slums: An application of multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

In this analysis we examine through an intersectionality lens how key social determinants of health (SDOH) are associated with health conditions among under-five children (<5y) residing in Nairobi slums, Kenya. We used cross-sectional data collected from Nairobi slums between June and November 2012 to explore how multiple interactions of SDoH shape health inequalities in slums. We applied multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) approach. We constructed intersectional strata for each health condition from combinations of significant SDoH obtained using univariate analyses. We then estimated the intersectional effects of health condition in a series of MAIHDA logistic regression models distinguishing between additive and interaction effects. We quantified discriminatory accuracy (DA) of the intersectional strata by means of the variance partitioning coefficient (VPC) and the area under the receiver operating characteristic curve (AUC-ROC). The total participants were 2,199 <5y, with 120 records (5.5%) dropped because health conditions were recorded as "not applicable". The main outcome variables were three health conditions: 1) whether a child had diarrhea or not, 2) whether a child had fever or not, and 3) whether a child had cough or not in the previous two weeks. We found non-significant intersectional effects for each health condition. The head of household ethnic group was significantly associated with each health condition. We found good DA for diarrhea (VPC = 9.0%, AUC-ROC = 76.6%) an indication of large intersectional effects. However, fever (VPC = 1.9%, AUC-ROC = 66.3%) and cough (VPC = 0.5%, AUC-ROC = 61.8%) had weak DA indicating existence of small intersectional effects. Our study shows pathways for SDoH that affect diarrhea, cough, and fever for <5y living in slums are multiplicative and shared. The findings show that <5y from Luo and Luhya ethnic groups, recent migrants (less than 2 years), and households experiencing CHE are more likely to face worse health outcomes. We recommend relevant stakeholders to develop strategies aimed at identifying these groups for targeted proportionate universalism based on the level of their need.

Understanding intersectional inequality in access to primary care providers using multilevel analysis of individual heterogeneity and discriminatory accuracy

BACKGROUND

Despite the Canadian healthcare system's commitment to equity, evidence for disparate access to primary care (PC) providers exists across individual social identities/positions. Intersectionality allows us to reflect the realities of how social power shapes healthcare experiences at an individual's interdependent and intersecting social identities/positions. The objectives of this study were to determine: (1) the extent to which intersections can be used classify those who had/did not have a PC provider; (2) the degree to which each social identity/position contributes to the ability to classify individuals as having a PC provider; and (3) predicted probabilities of having a PC provider for each intersection.

METHODS AND FINDINGS

Using national cross-sectional data from 241,445 individuals in Canada aged ≥18, we constructed 320 intersections along the dimensions of gender, age, immigration status, race, and income to examine the outcome of whether one had a PC provider. Multilevel analysis of individual heterogeneity and discriminatory accuracy, a multi-level model using individual-level data, was employed to address intersectional objectives. An intra-class correlation coefficient (ICC) of 23% (95%CI: 21-26%) suggests that these intersections could, to a very good extent, explain individual variation in the outcome, with age playing the largest role. Not all between-intersection variance in this outcome could be explained by additive effects of dimensions (remaining ICC: 6%; 95%CI: 2-16%). The highest intersectional predicted probability existed for established immigrant, older South Asian women with high income. The lowest intersectional predicted probability existed for recently immigrated, young, Black men with low income.

CONCLUSIONS

Despite a "universal" healthcare system, our analysis demonstrated a substantial amount of inequity in primary care across intersections of gender, age, immigration status, race, and income.

Non-response bias in the analysis of the association between mental health and the urban environment: a cross-sectional study in Brussels, Belgium

BACKGROUND

This paper aims at analysing the impact of partial non-response in the association between urban environment and mental health in Brussels. The potential threats of the partial non-response are biases in survey estimates and statistics. The effect of non-response on statistical associations is often overlooked and evidence in the research literature is lacking.

METHODS

Data from the Belgian Health Interview Survey 2008 and 2013 were used. The association between non-response and potential determinants was explored through logistic regressions.

RESULTS

Participants with low income, low educational levels, lower or higher age or in households with children were less likely to respond. When adjusting for socio-economic variables, non-response was higher in areas which are less vegetated, more polluted or more urbanised. Because the determinants of non-response and depressive disorders were similar, it is reasonable to assume that there will be more people with mental health problems among the non-respondents. And because more non-responses were found in low vegetation areas, the protective association between green spaces and mental health may be underestimated.

CONCLUSION

Our capacity to measure the association between the urban environment and health is affected by non-response in surveys. The non-random spatial and socio-economic distribution of this bias affects the research findings.

Educational Inequalities at the Intersection of Multiple Social Categories: An Introduction and Systematic Review of the Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) Approach

Intersectional approaches have become increasingly important for explaining educational inequalities because they help to improve our understanding of how individual experiences are shaped by simultaneous membership in multiple social categories that are associated with interconnected systems of power, privilege, and oppression. For years, there has been a call in psychological and educational research for quantitative approaches that can account for the intersection of multiple social categories. The present paper introduces the Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) approach, a novel intersectional approach from epidemiology, to study educational inequalities. The MAIHDA approach uses a multilevel model as the statistical framework to define intersectional strata that represent individuals’ membership in multiple social categories. By partitioning the variance within and between intersectional strata, the MAIHDA approach allows identifying intersectional effects at the strata level as well as obtaining information on the discriminatory accuracy of these strata for predicting individual educational outcomes. Compared to conventional quantitative intersectional approaches, MAIHDA analyses have several advantages, including better scalability for higher dimensions, model parsimony, and precision-weighted estimates of strata with small sample sizes. We provide a systematic review of its past application and illustrate its use by analyzing inequalities in reading achievement across 40 unique intersectional strata (combining the social categories of gender, immigrant background, parental education, and parental occupational status) using data from 15-year-old students in Germany (N = 5451). We conclude that the MAIHDA approach is a valuable intersectional tool to study inequalities in educational contexts.

Investigating people's attitudes towards participating in longitudinal health research: an intersectionality-informed perspective

BACKGROUND

Increasing evidence suggests that participation proportions in longitudinal health research vary according to sex/gender, age, social class, or migration status. Intersectionality scholarship purports that such social categories cannot be understood in isolation and makes visible the co-dependent nature of the social determinants of health and illness. This paper uses an intersectionality-informed approach in order to expand the understanding of why people participate in health research, and the impact of intersecting social structures and experiences on these attitudes.

METHODS

A sample of 80 respondents who had previously either accepted or declined an invitation to participate in the German National Cohort (NAKO) participated in our interview study. Interviews were semi-structured and contained both narrative elements and more structured probes. Data analysis proceeded in two steps: first, the entire data set was analysed thematically (separately for participants and non-participants); second, key themes were compared across self-reported sex/gender, age group and migration status to identify differences and commonalities.

RESULTS

Respondents' attitudes towards study participation can be categorised into four themes: wanting to make a contribution, seeking personalised health information, excitement and feeling chosen, and seeking social recognition. Besides citing logistical challenges, non-participants narrated adverse experiences with or attitudes towards science and the healthcare system that deterred them from participating. A range of social experiences and cultural value systems shaped such attitudes; in particular, this includes the cultural authority of science as an arbiter of social questions, transgressing social categories and experiences of marginalisation. Care responsibilities, predominantly borne by female respondents, also impacted upon the decision to take part in NAKO.

DISCUSSION

Our findings suggest that for participants, health research constitutes a site of distinction in the sense of making a difference and being distinct or distinguishable, whereas non-participants inhabited an orientation towards science that reflected their subjective marginalisation through science. No clear relationship can thereby be presumed between social location and a particular attitude towards study participation; rather, such attitudes transgress and challenge categorical boundaries. This challenges the understanding of particular populations as more or less disadvantaged, or as more or less inclined to participate in health research.

Intersectionality-Informed Sex/Gender-Sensitivity in Public Health Monitoring and Reporting (PHMR): A Case Study Assessing Stratification on an "Intersectional Gender-Score"

To date, PHMR has often relied on male/female stratification, but rarely considers the complex, intersecting social positions of men and women in describing the prevalence of health and disease. Stratification on an Intersectional Gender-Score (IG-Score), which is based on a variety of social covariables, would allow comparison of the prevalence of individuals who share the same complex intersectional profile (IG-Score). The cross-sectional case study was based on the German Socio-Economic Panel 2017 (n = 23,269 age 18+). After stratification, covariable-balance within the total sample and IG-Score-subgroups was assessed by standardized mean differences. Prevalence of self-rated health, mental distress, depression and hypertension was compared in men and women. In the IG-Score-subgroup with highest proportion of males and lowest probability of falling into the 'woman'-category, most individuals were in full-time employment. The IG-Score-subgroup with highest proportion of women and highest probability of falling into the 'woman'-category was characterized by part-time/occasional employment, housewife/-husband, and maternity/parental leave. Gender differences in prevalence of health indicators remained within the male-dominated IG-Score-subgroup, whereas the same prevalence of depression and self-rated health was observed for men and women constituting the female-dominated IG-Score-subgroup. These results might indicate that sex/gender differences of depression and self-rated health could be interpreted against the background of gender associated processes. In summary, the proposed procedure allows comparison of prevalence of health indicators conditional on men and women sharing the same complex intersectional profile.

The effects of social determinants on children's health outcomes in Bangladesh slums through an intersectionality lens: An application of multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

Empirical evidence suggests that the health outcomes of children living in slums are poorer than those living in non-slums and other urban areas. Improving health especially among children under five years old (U5y) living in slums, requires a better understanding of the social determinants of health (SDoH) that drive their health outcomes. Therefore, we aim to investigate how SDoH collectively affects health outcomes of U5y living in Bangladesh slums through an intersectionality lens. We used data from the most recent national Urban Health Survey (UHS) 2013 covering urban populations in Dhaka, Chittagong, Khulna, Rajshahi, Barisal, Sylhet, and Rangpur divisions. We applied multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) to estimate the Discriminatory Accuracy (DA) of the intersectional effects estimates using Variance Partition Coefficient (VPC) and the Area Under the Receiver Operating Characteristic Curve (AUC-ROC). We also assessed the Proportional Change in Variance (PCV) to calculate intersectional effects. We considered three health outcomes: cough, fever, and acute respiratory infections (ARI) in U5y.We found a low DA for cough (VPC = 0.77%, AUC-ROC = 61.90%), fever (VPC = 0.87%, AUC-ROC = 61.89%) and ARI (VPC = 1.32%, AUC-ROC = 66.36%) of intersectional strata suggesting that SDoH considered do not collectively differentiate U5y with a health outcome from those with and without a health outcome. The PCV for cough (85.90%), fever (78.42%) and ARI (69.77%) indicates the existence of moderate intersectional effects. We also found that SDoH factors such as slum location, mother's employment, age of household head, and household's garbage disposal system are associated with U5y health outcomes. The variables used in this analysis have low ability to distinguish between those with and without health outcomes. However, the existence of moderate intersectional effect estimates indicates that U5y in some social groups have worse health outcomes compared to others. Therefore, policymakers need to consider different social groups when designing intervention policies aimed to improve U5y health outcomes in Bangladesh slums.

CART-analysis embedded in social theory: A case study comparing quantitative data analysis strategies for intersectionality-based public health monitoring within and beyond the binaries

Implementation of an intersectionality perspective into quantitative health research might support the process of unravelling complex socio-cultural and economic power relations which underly and shape patterns of health and disease within populations. Intersectionality-informed health monitoring and health reporting integrating a sex/gender-sensitive perspective could serve as a fertile ground to strengthen the essential function of health reporting to support political decision-making. We aimed at the integration of social theory into quantitative data analysis by taking into consideration 4 distinct central sex/gender theoretical concepts in health research. We developed and tested an intersectionality-based, sex/gender-sensitive strategy comparing 5 distinct models based on different combinations of the binary sex/gender variable, socio-cultural and economic variables (defined from an intersectionality perspective) as well as solution-linked sex/gender variables. We used CART-analysis as a quantitative, non-parametric, exploratory method to detect subgroups with high prevalence of frequent mental distress (FMD). Analyses were based on data from a National Health Telephone Interview Survey conducted in Germany. Depending on model and detected subgroup of our comparative approach, prevalence of FMD ranged between approximately 5 %-25%. Within the model including the binary sex/gender variable, socio-cultural and economic variables, sex/gender turned out to be the most important attribute. Comparing the models which included solution-linked sex/gender variables to the model not including these variables illustrated that the CART-algorithm was able to detect subgroups with the same prevalence of FMD, but with approximately 14% as opposed to 4.5% of the study population being affected. For these models, social support served as the primary splitting variable and not the binary sex/gender variable. Including or not including the binary sex/gender variable in the models with the solution-linked variables did not make a substantial difference. Embedding CART-analysis in social theory might have the potential to further sex/gender sensitivity in health reporting and might support decision-making when considering the allocation of health-related interventions.