Human papillomavirus vaccination disparities among U.S. college students: An intersectional Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA)

Clarifications on the intersectional MAIHDA approach: A conceptual guide and response to Wilkes and Karimi (2024)

Intersectional Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) has been welcomed as a new gold standard for quantitative evaluation of intersectional inequalities, and it is being rapidly adopted across the health and social sciences. In their commentary "What does the MAIHDA method explain?", Wilkes and Karimi (2024) raise methodological concerns with this approach, leading them to advocate for the continued use of conventional single-level linear regression models with fixed-effects interaction parameters for quantitative intersectional analysis. In this response, we systematically address these concerns, and ultimately find them to be unfounded, arising from a series of subtle but important misunderstandings of the MAIHDA approach and literature. Since readers new to MAIHDA may share confusion on these points, we take this opportunity to provide clarifications. Our response is organized around four important clarifications: (1) At what level are the additive main effect variables defined in intersectional MAIHDA models? (2) Do MAIHDA models have problems with collinearity? (3) Why does the Variance Partitioning Coefficient (VPC) tend to be small, and the Proportional Change in Variance (PCV) tend to be large in MAIHDA? and (4) What are the goals of MAIHDA analysis?

A tutorial for conducting intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (I-MAIHDA) is an innovative approach for investigating inequalities, including intersectional inequalities in health, disease, psychosocial, socioeconomic, and other outcomes. I-MAIHDA and related MAIHDA approaches have conceptual and methodological advantages over conventional single-level regression analysis. By enabling the study of inequalities produced by numerous interlocking systems of marginalization and oppression, and by addressing many of the limitations of studying interactions in conventional analyses, intersectional MAIHDA provides a valuable analytical tool in social epidemiology, health psychology, precision medicine and public health, environmental justice, and beyond. The approach allows for estimation of average differences between intersectional strata (stratum inequalities), in-depth exploration of interaction effects, as well as decomposition of the total individual variation (heterogeneity) in individual outcomes within and between strata. Specific advice for conducting and interpreting MAIHDA models has been scattered across a burgeoning literature. We consolidate this knowledge into an accessible conceptual and applied tutorial for studying both continuous and binary individual outcomes. We emphasize I-MAIHDA in our illustration, however this tutorial is also informative for understanding related approaches, such as multicategorical MAIHDA, which has been proposed for use in clinical research and beyond. The tutorial will support readers who wish to perform their own analyses and those interested in expanding their understanding of the approach. To demonstrate the methodology, we provide step-by-step analytical advice and present an illustrative health application using simulated data. We provide the data and syntax to replicate all our analyses.

Extending intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) to study individual longitudinal trajectories, with application to mental health in the UK

The intersectional Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) approach is gaining prominence in health sciences and beyond, as a robust quantitative method for identifying intersectional inequalities in a range of individual outcomes. However, it has so far not been applied to longitudinal data, despite the availability of such data, and growing recognition that intersectional social processes and determinants are not static, unchanging phenomena. Drawing on intersectionality and life course theories, we develop a longitudinal version of the intersectional MAIHDA approach, allowing the analysis not just of intersectional inequalities in static individual differences, but also of life course trajectories. We discuss the conceptualization of intersectional groups in this context: how they are changeable over the life course, appropriate treatment of generational differences, and relevance of the age-period-cohort identification problem. We illustrate the approach with a study of mental health using United Kingdom Household Longitudinal Study data (2009-2021). The results reveal important differences in trajectories between generations and intersectional strata, and show that trajectories are partly multiplicative but mostly additive in their intersectional inequalities. This article provides an important and much needed methodological contribution, enabling rigorous quantitative, longitudinal, intersectional analyses in social epidemiology and beyond.

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.

Adverse Childhood Experiences and Sexual Orientation: An Intersectional Analysis of Nationally Representative Data

INTRODUCTION

This study compared the prevalence of adverse childhood experiences across intersections of sexual orientation, gender, race/ethnicity, and economic status.

METHODS

Data collected between 1994 and 2018 from 12,519 participants in the National Longitudinal Study of Adolescent to Adult Health were analyzed in 2023 to generate adverse childhood experience prevalence estimates. Unadjusted 1-way ANOVAs and multivariate regressions were performed to compare differences in independent and cumulative adversity measures by sexual orientation, gender, race/ethnicity, and poverty status. A multilevel analysis of individual heterogeneity and discriminatory accuracy was conducted to estimate adversity scores across 24 groups that were stratified by sexual orientation, gender, race/ethnicity, and poverty status.

RESULTS

Adolescents with same-sex attractions and adults who identified with a sexual minority group reported more adverse childhood experiences overall than straight participants, although associations varied by type of adversity. Strikingly, adversity scores were higher among White youth with same-sex attractions than among Black youth with same-sex attractions, among more economically advantaged bisexual adults than among poorer ones, and among poor White participants than among poor Black and Hispanic participants, suggesting that the combination of disadvantaged and marginalized statuses does not necessarily correspond with greater childhood adversity. A multilevel analysis of individual heterogeneity and discriminatory accuracy interaction model showed that sexual orientation and poverty status contributed significant variance to cumulative adversity scores, whereas gender and race/ethnicity did not.

CONCLUSIONS

The results show that disparities in adverse experiences can be more fully and accurately represented when sexual orientation and other social identities are modeled as intersectional configurations. Given that adverse childhood experiences are linked to morbidity and mortality, the findings have salient implications for understanding health disparities that affect population subgroups.

Mapping socio-geographical disparities in the occurrence of teenage maternity in Colombia using multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

BACKGROUND

The prevalence of teenage pregnancy in Colombia is higher than the worldwide average. The identification of socio-geographical disparities might help to prioritize public health interventions.

AIM

To describe variation in the probability of teenage maternity across geopolitical departments and socio-geographical intersectional strata in Colombia.

METHODS

A cross-sectional study based on live birth certificates in Colombia. Teenage maternity was defined as a woman giving birth aged 19 or younger. Multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) was applied using multilevel Poisson and logistic regression. Two different approaches were used: (1) intersectional: using strata defined by the combination of health insurance, region, area of residency, and ethnicity as the second level (2) geographical: using geopolitical departments as the second level. Null, partial, and full models were obtained. General contextual effect (GCE) based on the variance partition coefficient (VPC) was considered as the measure of disparity. Proportional change in variance (PCV) was used to identify the contribution of each variable to the between-strata variation and to identify whether this variation, if any, was due to additive or interaction effects. Residuals were used to identify strata with potential higher-order interactions.

RESULTS

The prevalence of teenage mothers in Colombia was 18.30% (95% CI 18.20-18.40). The highest prevalence was observed in Vichada, 25.65% (95% CI: 23.71-27.78), and in the stratum containing mothers with Subsidized/Unaffiliated healthcare insurance, Mestizo, Rural area in the Caribbean region, 29.08% (95% CI 28.55-29.61). The VPC from the null model was 1.70% and 9.16% using the geographical and socio-geographical intersectional approaches, respectively. The higher PCV for the intersectional model was attributed to health insurance. Positive and negative interactions of effects were observed.

CONCLUSION

Disparities were observed between intersectional socio-geographical strata but not between geo-political departments. Our results indicate that if resources for prevention are limited, using an intersectional socio-geographical approach would be more effective than focusing on geopolitical departments especially when focusing resources on those groups which show the highest prevalence. MAIHDA could potentially be applied to many other health outcomes where resource decisions must be made.

A novel application of interrupted time series analysis to identify the impact of a primary health care reform on intersectional inequities in avoidable hospitalizations in the adult Swedish population

Primary health care (PHC) systems are a crucial instrument for achieving equitable population health, but there is little evidence of how PHC reforms impact equities in population health. In 2010, Sweden implemented a reform that promoted marketization and privatization of PHC. The present study uses a novel integration of intersectionality-informed and evaluative epidemiological analytical frameworks to disentangle the impact of the 2010 Swedish PHC reform on intersectional inequities in avoidable hospitalizations. The study population comprised the total Swedish population aged 18-85 years across 2001-2017, in total 129 million annual observations, for whom register data on sociodemographics and hospitalizations due to ambulatory care sensitive conditions were retrieved. Multilevel Analysis of Individual Heterogeneity and Discriminatory Analyses (MAIHDA) were run for the pre-reform (2001-2009) and post-reform (2010-2017) periods to provide a mapping of inequities. In addition, random effects estimates reflecting the discriminatory accuracy of intersectional strata were extracted from a series MAIHDAs run per year 2001-2017. The estimates were re-analyzed by Interrupted Time Series Analysis (ITSA), in order to identify the impact of the reform on measures of intersectional inequity in avoidable hospitalizations. The results point to a complex reconfiguration of social inequities following the reform. While the post-reform period showed a reduction in overall rates of avoidable hospitalizations and in age disparities, socioeconomic inequities in avoidable hospitalizations, as well as the importance of interactions between complex social positions, both increased. Socioeconomically disadvantaged groups born in the Nordic countries seem to have benefited the least from the reform. The study supports a greater attention to the potentially complex consequences that health reforms can have on inequities in health and health care, which may not be immediate apparent in conventional evaluations of either population-average outcomes, or by simple evaluations of equity impacts. Methodological approaches for evaluation of complex inequity impacts need further development.

Routine Vaccination Coverage in an Adolescent Transgender Population in a Large Tertiary Care Center in the United States

INTRODUCTION

To address healthcare disparities among transgender and gender diverse (TGD) adolescents, we examined vaccination coverage in those receiving gender-affirming care.

METHOD

Our study analyzed de-identified data (2013-2022) from a tertiary care clinic. Comparing vaccination rates of 203 TGD adolescent patients to age-matched peers in New York State using CDC National Immunization Survey-Teen data.

RESULTS

We found TGD patients had similar vaccination coverage to the general adolescent population. Notably, TGD patients had significantly higher up-to-date human papillomavirus vaccination coverage (76.4%, CI=69.9, 82.0) than the NYS adolescent population (60.5%, CI=53.0, 67.5).

DISCUSSION

This suggests TGD adolescents at gender-affirming clinics maintain or exceed vaccination rates of their cisgender counterparts. Future research should explore vaccination rates for TGD individuals without access to gender-affirming care.

Overcoming combination fatigue: Addressing high-dimensional effect measure modification and interaction in clinical, biomedical, and epidemiologic research using multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA)

Growing interest in precision medicine, gene-environment interactions, health equity, expanding diversity in research, and the generalizability results, requires researchers to evaluate how the effects of treatments or exposures differ across numerous subgroups. Evaluating combination complexity, in the form of effect measure modification and interaction, is therefore a common study aim in the biomedical, clinical, and epidemiologic sciences. There is also substantial interest in expanding the combinations of factors analyzed to include complex treatment protocols (e.g., multiple study arms or factorial randomization), comorbid medical conditions or risk factors, and sociodemographic and other subgroup identifiers. However, expanding the number of subgroup category combinations creates combination fatigue problems, including concerns over small sample size, reduced power, multiple testing, spurious results, and design and analytic complexity. Creative new approaches for managing combination fatigue and evaluating high-dimensional effect measure modification and interaction are needed. Intersectional MAIHDA (multilevel analysis of individual heterogeneity and discriminatory accuracy) has already attracted substantial interest in social epidemiology, and has been hailed as the new gold standard for investigating health inequities across complex intersections of social identity. Leveraging the inherent advantages of multilevel models, a more general multicategorical MAIHDA can be used to study statistical interactions and predict effects across high-dimensional combinations of conditions, with important advantages over alternative approaches. Though it has primarily been used thus far as an analytic approach, MAIHDA should also be used as a framework for study design. In this article, I introduce MAIHDA to the broader health sciences research community, discuss its advantages over conventional approaches, and provide an overview of potential applications in clinical, biomedical, and epidemiologic research.

An Intersectional Approach to Quantifying the Impact of Geographic Remoteness and Health Disparities on Quality-Adjusted Life Expectancy: Application to Australia

OBJECTIVES

An intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) is a novel method for exploring the interaction between sociodemographic characteristics that affect health outcomes. This study explores the interaction between geographic remoteness and socioeconomic status on health outcomes in Australia from an intersectional perspective.

METHODS

Data from a cross-sectional survey were matched with data from the Australian Bureau of Statistics and the Australian Institute of Health and Welfare. To explore the effect of health-related quality of life on life expectancy, quality-adjusted life expectancy (QALE) was estimated through applying utility values derived from the EQ-5D-5L to life table data from the Australian Bureau of Statistics. The effect of geographic remoteness on QALE was quantified using multivariable linear regression. An intersectional MAIHDA was performed to explore differences in mean QALE across strata formed by intersections of age, sex, and Socioeconomic Indexes for Areas score.

RESULTS

Based on multivariable linear modeling, QALE declined significantly with increasing remoteness (inner regional, -1.0 years [undiscounted]; remote/very remote, -3.3 years [undiscounted]) (P < .001). In contrast, life expectancy was only significantly different between participants in remote/very remote areas and major cities (β-coefficient, -2.4; 95% CI -4.4 to -0.4; P = .016). No intersectional interaction effects between strata on QALE were found in the MAIHDA.

CONCLUSIONS

QALE has considerable value as a metric for exploring disparities in health outcomes. Given that no intersectional interactions were identified, our findings support broad interventions that target the underlying social determinants of health appropriately reduce disparities versus interventions targeting intersectional interactions.

Intersectional Prevalence of Suicide Ideation, Plan, and Attempt Based on Gender, Sexual Orientation, Race and Ethnicity, and Rurality

Importance

Suicidal thoughts and behaviors (STBs) are major public health problems, and some social groups experience disproportionate STB burden. Studies assessing STB inequities for single identities (eg, gender or sexual orientation) cannot evaluate intersectional differences and do not reflect that the causes of inequities are due to structural-level (vs individual-level) processes.

Objective

To examine differences in STB prevalence at the intersection of gender, sexual orientation, race and ethnicity, and rurality.

Design, Setting, and Participants

This cross-sectional study used adult data from the 2015-2019 National Survey on Drug Use and Health (NSDUH), a population-based sample of noninstitutionalized US civilians. Data were analyzed from July 2022 to March 2023.

Main Outcomes and Measures

Outcomes included past-year suicide ideation, plan, and attempt, each assessed with a single question developed for the NSDUH. Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) models were estimated, in which participants were nested within social strata defined by all combinations of gender, sexual orientation, race and ethnicity, and rurality; outcome prevalence estimates were obtained for each social stratum. Social strata were conceptualized as proxies for exposure to structural forms of discrimination that contribute to health advantages or disadvantages (eg, sexism, racism).

Results

The analytic sample included 189 800 adults, of whom 46.5% were men; 53.5%, women; 4.8%, bisexual; 93.0%, heterosexual; 2.2%, lesbian or gay; 18.8%, Hispanic; 13.9%, non-Hispanic Black; and 67.2%, non-Hispanic White. A total of 44.6% were from large metropolitan counties; 35.5%, small metropolitan counties; and 19.9%, nonmetropolitan counties. There was a complex social patterning of STB prevalence that varied across social strata and was indicative of a disproportionate STB burden among multiply marginalized participants. Specifically, the highest estimated STB prevalence was observed among Hispanic (suicide ideation: 18.1%; 95% credible interval [CrI], 13.5%-24.3%) and non-Hispanic Black (suicide plan: 7.9% [95% CrI, 4.5%-12.1%]; suicide attempt: 3.3% [95% CrI, 1.4%-6.2%]) bisexual women in nonmetropolitan counties.

Conclusions and Relevance

In this cross-sectional study, intersectional exploratory analyses revealed that STB prevalence was highest among social strata including multiply marginalized individuals (eg, Hispanic and non-Hispanic Black bisexual women) residing in more rural counties. The findings suggest that considering and intervening in both individual-level (eg, psychiatric disorders) and structural-level (eg, structural discrimination) processes may enhance suicide prevention and equity efforts.

The intersections of socioeconomic position, gender, race/ethnicity and nationality in relation to oral conditions among American adults

OBJECTIVE

The objective of the study was to evaluate how gender, socioeconomic position (SEP), race/ethnicity and nationality intersect to structure social inequalities in adult oral health among American adults.

METHODS

Data from adults aged 20 years or over who participated in the National Health and Nutrition Examination Survey (NHANES) 2009-2018 were analysed. The outcomes were poor self-rated oral health and edentulism among all adults (n = 24 541 and 21 446 participants, respectively) and untreated caries and periodontitis among dentate adults (n = 16 483 and 9829 participants, respectively). A multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) was conducted for each outcome, by nesting individuals within 48 intersectional strata defined as combinations of gender, SEP, race/ethnicity and nationality. Intersectional measures included the variance partition coefficient (VPC), the proportional change in variance (PCV) and predicted excess probability due to interaction.

RESULTS

Substantial social inequalities in the prevalence of oral conditions among adults were found, which were characterized by high between-stratum heterogeneity and outcome specificity. The VPCs of the simple intersectional model showed that 9.4%-12.7% of the total variance in the presentation of oral conditions was attributed to between-stratum differences. In addition, the PCVs from the simple intersectional model to the intersectional interaction model showed that 84.1%-97.1% of the stratum-level variance in the presentation of oral conditions was attributed to the additive effects of gender, SEP, race/ethnicity and nationality. The point estimates of the predictions for some intersectional strata were suggestive of an intersectional interaction effect. However, the 95% credible intervals were very wide and the estimations inconclusive.

CONCLUSIONS

This analysis highlights the value of the intersectionality framework to understand heterogeneity in social inequalities in oral health. These inequalities were mainly due to the additive effect of the social identities defining the intersectional strata, with no evidence of interaction effects.

Predictor of HPV Vaccination Uptake among Foreign-Born College Students in the U.S.: An Exploration of the Role of Acculturation and the Health Belief Model

OBJECTIVE

to measure the HPV vaccination rate and knowledge about HPV and its vaccine among foreign-born college students; additionally, to measure the effect of acculturation and HBM constructs on the HPV vaccination behavior among foreign-born college students.

METHODS

a cross-sectional design with a non-probability sample of foreign-born college students was collected via a web-based self-administered survey that measured the HPV vaccination rate, assessed knowledge about HPV and its vaccine, and evaluated the effect of acculturation and HBM constructs on HPV vaccination behavior among foreign-born college students.

RESULTS

Foreign-born college students had moderate knowledge about HPV and the HPV vaccine, and about 63% were HPV-vaccinated. Perceived susceptibility, perceived barriers, and cues to action were significantly associated with the HPV vaccination behavior, while knowledge levels about HPV and the HPV vaccine and acculturation levels were not significantly associated with the HPV vaccination behavior of foreign-born college students.

CONCLUSIONS

The current study shows a moderate vaccination rate and moderate knowledge about HPV and its vaccine among foreign-born college students. Additionally, vaccination campaigns need to increase awareness about the susceptibility to acquiring HPV and minimize the barriers to receiving the vaccine to increase the HPV vaccination rate among foreign-born college students.

Multilevel modelling for measuring interaction of effects between multiple categorical variables: An illustrative application using risk factors for preeclampsia

BACKGROUND

Measuring multiple and higher-order interaction effects between multiple categorical variables proves challenging.

OBJECTIVES

To illustrate a multilevel modelling approach to studying complex interactions.

METHODS

We apply a two-level random-intercept linear regression to a binary outcome for individuals (level-1) nested within strata (level-2) defined by all observed combinations of multiple categorical exposure variables. As a pedagogic application, we analyse 36 strata defined by five risk factors of preeclampsia (parity, previous preeclampsia, chronic hypertension, multiple pregnancies, body mass index category) among 652,603 women in the Swedish Medical Birth Registry between 2002 and 2010.

RESULTS

The absolute risk of preeclampsia was 4% but was predicted to vary from 1% to 44% across strata. The stratum discriminatory accuracy was 30% according to the variance partition coefficient (VPC) and 0.73 according to the area under the receiver operating characteristic curve (AUC). While the risk heterogeneity across strata was primarily due to the main effects of the categories defining the strata, 5% of the variation was attributable to their two- and higher-way interaction effects. One stratum presented a positive interaction, and two strata presented negative interaction.

CONCLUSIONS

Multilevel modelling is an innovative tool for identifying and analysing higher-order interaction effects. Further work is needed to explore how this approach can best be applied to making causal inferences.