If you look at many recent NIH funding opportunities focused on interventions to address health disparities, you may see language like this:

• Proposed studies must address “determinants at two or more socio-ecological levels” (NOT-OD-24-058).
• Proposed interventions should “seek to foster change in the environment (physical, built or sociocultural) at the organizational and/or community level” (PAR-23-066).
• Projects that “intervene solely at the individual/family level and not on SDOH [social determinants of health] as conceptualized by the NIH” are not responsive (RFA-NR-24-004).
• Projects that “use an individual-level randomized trial design” are not responsive (PAR-24-053).

Such stipulations show that NIH and the broader research community recognize the need for interventions that move beyond individual factors to modify adverse social determinants that create and perpetuate health disparities. (You can reference this article for more information about using multilevel interventions to address health disparities.) As this recognition grows, so does the need for appropriate study designs and analytic methods to evaluate these complex interventions.

ODP-Sponsored Journal Supplement on Methods

To address this need, ODP sponsored a supplemental issue in the journal Prevention Science entitled “Design and Analytic Methods to Evaluate Multilevel Interventions to Reduce Health Disparities.” Our purpose in publishing the supplemental issue was two-fold: (1) bring together a collection of state-of-the-science papers on relevant methods in a single issue, and (2) counter a common perception that it is impossible to use rigorous clinical trial methods for community-based research in marginalized, minoritized, or underserved populations. We believe that the collection of papers in the supplemental issue provides strong counterevidence to this view.

As Dr. Melody Goodman and I discuss in our accompanying commentary in the issue, these papers offer a mix of new approaches to the design and analysis of multilevel interventions and the application of existing methods, with an added focus on health equity. Papers cover the continuum of multilevel intervention research, from intervention development and sample size estimation to analyses of intervention effects by subgroups, using a variety of randomized trial approaches, including parallel group and stepped wedge trial designs and optimization designs.

Additionally, the clinical trials described in the papers address a wide range of health conditions or outcomes (e.g., mental health, reproductive health, hypertension, COVID-19), settings (e.g., health care, long-term care, workplace, school), and populations (e.g., racial/ethnic minoritized groups, people with lower incomes, rural residents). We hope that the breadth and depth of material covered in this issue will provide a useful resource for investigators across disciplines and topic areas.

Commitment to Improving Research Methods

Let me close by drawing attention to two related resources from ODP.

NIH Research Methods Resources: This website provides information, key references, sample size calculators, and other resources for research designs commonly used in clinical trials, including those designed to reduce health disparities. ODP has already incorporated the methods reported in the supplemental issue by Hughes et al. (2024) into the sample size calculator for stepped wedge group-randomized trials to allow investigators to plan studies in which the intervention effect may vary over time. The website also provides information on parallel group- or cluster-randomized trials, individually randomized group-treatment trials, and group or cluster regression discontinuity designs.

Methods: Mind the Gap Webinar Series: This monthly series provides a forum for presentation and discussion of current methodological research relevant to clinical trials supported by NIH, including those designed to reduce health disparities. Recent webinars include presentations on several of the papers included in the supplemental issue. For example:

• Drs. Linda Collins and Jillian Strayhorn presented on Optimizing Interventions for Equitability: Some Initial Ideas. 
• Dr. Jim Hughes presented on Robust Inference for Stepped Wedge Designs. 
• Drs. Keith Muller and Deborah Glueck presented on Choosing Sample Sizes for Multilevel and Longitudinal Studies Analyzed with Linear Mixed Models.

Future webinars will continue to feature talks by the authors of papers from the supplemental issue. I encourage you to subscribe to our listserv to stay up to date about upcoming webinars and activities from ODP.

I hope that this supplemental issue, together with the NIH Research Methods Resources website and the Methods: Mind the Gap webinars, will help investigators use state-of-the-science methods to design and analyze their trials to evaluate multilevel interventions. We won’t advance the science on reducing health disparities if we don’t use the best methods available.