Dataset Information

Quantitative methods for descriptive intersectional analysis with binary health outcomes

ABSTRACT: Intersectionality recognizes that in the context of sociohistorically shaped structural power relations, an individual's multiple social positions or identities (e.g., gender, ethnicity) can interact to affect health-related outcomes. Despite limited methodological guidance, intersectionality frameworks have increasingly been incorporated into epidemiological studies, both to describe health disparities and to examine their causes. This study aimed to advance methods in intersectional estimation of binary outcomes in descriptive health disparities research through evaluation of 7 potentially intersectional data analysis methods: cross-classification, regression with interactions, multilevel analysis of individual heterogeneity (MAIHDA), and decision trees (CART, CTree, CHAID, random forest). Accuracy of estimated intersection-specific outcome prevalence was evaluated across 192 intersections using simulated data scenarios. For comparison we included a non-intersectional main effects regression. We additionally assessed variable selection performance amongst decision trees. Example analyses using National Health and Nutrition Examination Study data illustrated differences in results between methods. At larger sample sizes, all methods except for CART performed better than non-intersectional main effects regression. In smaller samples, MAIHDA was the most accurate method but showed no advantage over main effects regression, while random forest, cross-classification, and saturated regression were the least accurate, and CTree and CHAID performed moderately well. CART performed poorly for estimation and variable selection. Sensitivity analyses examining the bias-variance tradeoff suggest MAIHDA as the preferred unbiased method for accurate estimation of high-dimensional intersections at smaller sample sizes. Larger sample sizes are more imperative for other methods. Results support the adoption of an intersectional approach to descriptive epidemiology. Highlights • Intersectionality may inform methods for describing health inequity.• Seven statistical and machine learning methods were evaluated for accuracy.• For binary health outcomes, method performance was quite sensitive to sample size.• MAIHDA was preferred for accurate intersection-specific estimation in small samples.• In larger samples most methods out-performed non-intersectional main effects.

SUBMITTER: Mahendran M

PROVIDER: S-EPMC8800141 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundMeta-analyses (MA) based on individual patient data (IPD) are regarded as the gold standard for meta-analyses and are becoming increasingly common, having several advantages over meta-analyses of summary statistics. These analyses are being undertaken in an increasing diversity of settings, often having a binary outcome. In a previous systematic review of articles published between 1999-2001, the statistical approach was seldom reported in sufficient detail, and the outcome was binary in 32% of the studies considered. Here, we explore statistical methods used for IPD-MA of binary outcomes only, a decade later.MethodsWe selected 56 articles, published in 2011 that presented results from an individual patient data meta-analysis. Of these, 26 considered a binary outcome. Here, we review 26 IPD-MA published during 2011 to consider: the goal of the study and reason for conducting an IPD-MA, whether they obtained all the data they sought, the approach used in their analysis, for instance, a two-stage or a one stage model, and the assumption of fixed or random effects. We also investigated how heterogeneity across studies was described and how studies investigated the effects of covariates.Results19 of the 26 IPD-MA used a one-stage approach. 9 IPD-MA used a one-stage random treatment-effect logistic regression model, allowing the treatment effect to vary across studies. Twelve IPD-MA presented some form of statistic to measure heterogeneity across studies, though these were usually calculated using two-stage approach. Subgroup analyses were undertaken in all IPD-MA that aimed to estimate a treatment effect or safety of a treatment,. Sixteen meta-analyses obtained 90% or more of the patients sought.ConclusionEvidence from this systematic review shows that the use of binary outcomes in assessing the effects of health care problems has increased, with random effects logistic regression the most common method of analysis. Methods are still often not reported in enough detail. Results also show that heterogeneity of treatment effects is discussed in most applications.

Project description:BackgroundGuided by intersectionality frameworks, researchers have documented health disparities at the intersection of multiple axes of social status and position, particularly race and ethnicity, gender, and sexual orientation. To advance from identifying to intervening in such intersectional health disparities, studies that examine the underlying mechanisms are required. Although much research demonstrates the negative health impacts of perceived discrimination along single axes, quantitative approaches to assessing the role of discrimination in generating intersectional health disparities remain in their infancy. Members of our team recently introduced the Intersectional Discrimination Index (InDI) to address this gap. The InDI comprises three measures of enacted (day-to-day and major) and anticipated discrimination. These attribution-free measures ask about experiences of mistreatment because of who you are. These measures show promise for intersectional health disparities research but require further validation across intersectional groups and languages. In addition, the proposal to remove attributions is controversial, and no direct comparison has ever been conducted.ObjectiveThis study aims to cognitively and psychometrically evaluate the InDI in English and Spanish and determine whether attributions should be included.MethodsThe study will draw on a preliminary validation data set and three original sequentially collected sources of data: qualitative cognitive interviews in English and Spanish with a sample purposively recruited across intersecting social status and position (gender, sexual orientation, race and ethnicity, socioeconomic status, age, and nativity); a Spanish quantitative survey (n=500; 250/500, 50% sexual and gender minorities); and an English quantitative survey (n=3000), with quota sampling by race and ethnicity (Black, Latino/a/x, and White), sexual or gender minority status, and gender.ResultsThe study was funded by the National Institute on Minority Health and Health Disparities in May 2021, and data collection began in July 2021.ConclusionsThe key deliverables of the study will be bilingual measures of anticipated, day-to-day, and major discrimination validated for multiple health disparity populations using qualitative, quantitative, and mixed methods.International registered report identifier (irrid)PRR1-10.2196/30987.

Project description:BackgroundThere is growing evidence suggesting that the emotional well-being of the public has been negatively affected in the past year. Consequently, demand for well-being support has increased. Although there is substantial empirical support for mental health apps that target diagnosed conditions, there is less research on emotional well-being apps. Among existing well-being apps, few studies have been conducted on apps that are based on lived experience and those that seek to enhance users' understanding of their emotional patterns. Thus, the acceptability of these novel apps requires further evaluation before upscaling.ObjectiveThis evaluation aims to describe the acceptability, engagement, and preliminary outcomes of using an app (Paradym) designed to promote emotional well-being and positive mental health.MethodsThis is a pre-post, mixed-methods, single-arm evaluation that is aggregated with digital analytics data. We anonymously collected real-world data on the demographics and well-being of the participants as well as the usability and acceptance of the app using validated questionnaires and open-ended questions. Participants tested the app for a minimum of 2 weeks before completing the follow-up measures. Google Analytics was used to record the level of app engagement. Chi-square and 2-tailed t tests were conducted to analyze quantitative data, and a thematic analysis approach was adopted for qualitative data.ResultsA total of 115 participants completed baseline questionnaires, of which 79.1% (91/115) users downloaded the app. The sample was diverse in terms of ethnicity, including 43.4% (50/115) people who self-identified as belonging to minority ethnic groups. Most of the participants were female (78/115, 67.8%) and between the ages of 18 and 25 years (39/115, 33.9%). A total of 34 app users who completed questionnaires at baseline and follow-up provided valuable feedback to inform the future directions of Paradym. Favorable themes emerged describing the app's content, functionality, and underlying principles. Although usability feedback varied across items, a considerable number of participants (22/34, 64%) found that the app was easy to use. Google Analytics revealed that at least 79% (27/34) of people used the app daily. On the basis of preliminary observations, app users experience increased mental well-being. Post hoc analyses indicated that the reduction in depression scores (t33=-2.16) and the increase in the well-being measures (t33=2.87) were statistically significant. No adverse events were reported during the follow-up period.ConclusionsThe findings of this evaluation are encouraging and document positive preliminary evidence for the Paradym app.

Dataset Information

Quantitative methods for descriptive intersectional analysis with binary health outcomes

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets