Transportability, the ability to maintain performance across populations, is a desirable property of markers of clinical outcomes. However, empirical findings indicate that markers often exhibit varying performances across populations. For prognostic markers that are advertised as predictive risk equations for an outcome of interest, oftentimes a form of updating is required when the equation is transported to populations with different outcome prevalences. Here, we revisit transportability of prognostic markers through the lens of the foundational framework of sufficient-component causes. We argue that transporting a marker "as is" implicitly assumes predictive values are transportable, whereas conventional prevalence-adjustment shifts the locus of transportability to accuracy metrics (sensitivity and specificity). Using a minimalist sufficient-cause framework that decomposes risk prediction into broad causal constituents, we show that both approaches rely on strong assumptions about the stability of cause distributions. Such a framework instead invites making transparent assumptions about how different causes vary across populations, leading to different transportability methods. For example, in the absence of any external information other than outcome prevalence, an impartial perspective can assume all causes are responsible for change in prevalence. This assumption leads to a new form of marker transportability method. Numerical experiments demonstrate that different transportability assumptions lead to varying degrees of information loss, depending on the distribution of causes across populations. A sufficient-component-cause perspective challenges common assumptions and practices for marker transportability, and results in novel transportability methods based on explicit assumptions on how different causes vary across populations.

Adolescents may experience multiple early adversities alongside mental and behavioral challenges whose inter-relationships are not well captured by traditional conceptual models. Network analysis examines the conditional associations among indicators, offering new insights into how these experiences and behaviors cluster and interconnect. Drawing on a nationally representative sample of U.S. adolescents, we estimated an Ising network of 26 items from the 2023 national Youth Risk Behavior Survey (YBRS, N = 20,103) across five domains: adverse childhood experiences (ACEs), victimization, suicidal behaviors, substance use, and violence. We examined node and bridge centrality and assessed network invariance through the Network Comparison Test by sex (female vs. male), sexual orientation (heterosexual vs. sexual minorities and bisexuals separately), and race/ethnicity, comparing White respondents to Black/African American, Hispanic/Latino, Asian, American Indian/Alaska Native, and Multiracial respondents. The network was stable with substantial overall connectivity. Three variables showed the highest centrality (as measured by node strength z-scores): suicidal ideation (1.8), caregiver emotional abuse (1.8), and hopelessness (1.7). The strongest between-domain connectors (bridge strength z-scores) were hopelessness (2.2), physical fighting (2.0), and gun carrying (1.6). Hopelessness emerged as the most important variable in the network. Caregiver emotional abuse was the most relevant node within the ACEs. Physical fights and gun carrying were key in connecting violence to other domains. Exploratory subgroup comparisons suggested broadly similar network structure and global strength across sex, sexual orientation, and most racial/ethnic comparisons, with the strongest evidence of structural differences for Black/African American versus White respondents. These findings offer new directions for future research by identifying potential indicators of risk or cascade processes that may inform hypotheses about whether the observed associations reflect causal pathways.

Covariate selection is a challenging part of estimating effects, whether accounting for confounding or improving external validity by addressing different distributions of effect measure modifiers between a study and target population. However, current recommendations on covariate selection for external validity analyses might be misleading. We investigate how directions of causal relationships between sample membership, types of effect measure modification, and collider bias impact sufficient adjustment sets for external validity analysis. We provide a simulated example to underpin the theoretical results and examine how impacts of collider bias depend on the colliders' associations, strength of interactions and sample size. When sample membership is caused by effect measure modifiers (e.g., when sociodemographics have an effect on the probability of trial participation), conditioning on effect measure modifiers (regardless of type) does not bias target population estimates. However, in the converse scenario where effect measure modifiers are caused by sample membership (e.g., under trial engagement effects), conditioning on some types of effect measure modifiers can be problematic. This is because sample membership can become an effect measure modifier by collider conditioning. We term adjustment sets that differ in their sufficiency according to the specific causal relationships between effect measure modifiers and sample membership "selection dependent" (s-dependent) sets. It is important to note, however, that weak interactions and collider-parent associations may not be sufficient to generate substantial collider bias. Nevertheless, we argue that researchers should consider the direction of the relationships between sample membership and effect measure modifiers to avoid inducing collider bias in external validity analyses.

In randomized trials, the per-protocol effect is defined as the effect of being assigned a treatment strategy and receiving treatment according to the assigned strategy; therefore, it encompasses the effect of the treatment strategy and any effect of assignment that does not operate through the received treatment. However, the per-protocol effect is sometimes interpreted as reflecting the effect of the treatment strategy itself, regardless of assignment. Here, we argue by example that this is not necessarily the case. We examine causal structures for randomized trials where these two causal estimands - the per-protocol effect and the effect of the treatment strategy - are not equal, and where their corresponding identifying observed data functionals are not the same, but both require information on assignment for identification. Our examples clarify the conceptual difference between these estimands and the conditions that guarantee their equality. Our examples also highlight that in some cases identification of these estimands requires information on assignment, even when assignment is randomized, unless one makes additional assumptions - informally, that assignment does not affect the outcome except through treatment (i.e., an exclusion-restriction assumption), and that assignment is not a confounder of the treatment-outcome association conditional on other variables in the analysis. These assumptions may not be plausible, particularly when the treatment assignment process in the trial differs materially from routine practice in outcome-relevant ways (e.g., "blinding" investigators and participants to treatment assignment). Our examination emphasizes the important role of assignment in defining and estimating causal effects in randomized trials, with implications for statistical analysis and the interpretation of trial results.