Unlocking the Power of REaL and SOGI Data: Lessons for Healthcare Professionals

I want to be REAL for a second.  

Most healthcare professionals do not understand the importance of REaL and SOGI data.  They don't understand the impact this information can have on healthcare transformation.  They are unsure if they want to learn another tech solution to capture this information, nor have they received the training to do it in a way that does not stigmatize them.  

This is not our fault.  Often, this information is not shared; it operates on a need-to-know basis.  I’ve been there; I was definitely not always in the know.  It's time to break it down.  I'm here to educate, inform, collaborate, and share the six lessons Ive learned in the past year about collecting race, ethnicity, language, disability, sexual orientation, and gender identity. 

I am not a healthcare data expert, but I have taken numerous data-related courses. And I am passionate about ensuring the underserved have access to quality healthcare. So, I went on a journey to learn all I could about collecting this data.  Since that decision, REAL/SOGI data has become consistent in my life.  

Let's jump in and discuss and share some resources. 

Lesson 1: Understanding the Impact

One of the first things you learn in curriculum development and change management is to start with the “why.”  When people understand the “why,” they are more open to accepting the information and change.  What is the why behind collecting REaL and SOGI data?

Over 20 years ago, the Institute of Medicine released a report, Unequal Treatment.  It was one of the first studies to demonstrate the existence of healthcare disparities, specifically among different racial and ethnic groups, but also for different genders and disabilities.  From that moment on, healthcare began a more visible journey of understanding these disparities.  

Collecting demographic data lets us  “see” where there are disparities.  Outside of academia, professionals finally had access to information showing the monumental gaps in the quality of care and outcomes for marginalized groups.   We were given insights that could drive our decision-making.  It opened the door for initiatives around access.  And it all began with collecting data on race, ethnicity, language, disability status, sexual orientation, and gender identity. 

This advancement has not been without challenges.  In fact, the last twenty years have been about trying new ideas to address these challenges.  

Lesson 2: Overcoming Challenges

 We could discuss challenges all day, but most people skim these newsletters 😂.  I'm only going to give you three.  In the comments, let me know your organization's challenges in this part of the health equity journey.

Lack of Standardization

Once I knew part of my role was building a strategy to collect REaL/SOGI data, I always looked for evidence-based frameworks to collect the data.  That one search took me down a rabbit hole. 

Currently, there is no widely accepted way to collect, store, reconcile, and use REaL and SOGI data.  When something is not standard practice, everyone begins putting their spin on it, which limits collaboration, interoperability), and innovation.  The only agreement is using OMB categorization, which is widely used for federal surveys.  This is only a starting point.  For healthcare, this categorization limits your insights.   I advise creating a cross-functional work group to decide on how this work is prioritized and creating a road map for the future.  OMB categories are going to change, and research is still evolving.   

BTW…..notice I did not mention gender identity and sexual orientation at all in the above paragraph.  That is not by mistake.  While there is guidance for race/ethnicity, collecting SOGI (that does not stigmatize and offers various categories) is relatively new.  The federal government released best practices for collecting this data on federal surveys.  It is a starting point.  Other valuable resources are listed in the resource section. 

Interoperability

We’ve got a problem we’ve had for a while.  Throughout a person’s lifespan, they will see many doctors in different locations through varied health insurance.  We know that information needs to follow the person wherever they go, but this is not the US healthcare experience.  An article in Communication Medicine very clearly stated, “This lack of interoperability within and across hospital systems, laboratories, public health programs, physicians’ offices, and regulatory and research data resources hinders rapid improvements in medical treatment, public health, decision-making, and research (Ana Szarfman, et al., 2022).”  

I recommend reviewing the work happening now by the real experts.  The Office for the National Coordinator of Health IT has a path forward for interoperability.  There are five elements.  Guess what? We already discussed number 2, standards (see, I do know just a little bit). As your organization makes decisions about interoperability, you will run into the next challenge….

“To aggregate? To disaggregate?”

Aggregation vs Disaggregation

Oooooh, let me tell you, I felt this challenge.   If you are new to healthcare data, I can easily explain this concept using an example.

In this case, let's talk about Native Americans in the US.  Now, remember, our first challenge was around standards and OMB categorization.   OMB categorizations allow only one option for Native Americans and Alaska Natives.  Every person who selects this option, no matter their tribal affiliation, gets lumped into this category, i.e., aggregation.  Now, there are over 570 different federally recognized tribes in the US, and their healthcare experience, while similar, will also have differences.  For example, the experience of a Native American tribe in Oregon is different from that of a tribe in South Carolina.  They are not a monolith.  Put this in your memory bank: none of the race/ethnicity boxes that we put people into are monoliths.

So which is more important, aggregating or disaggregating data?  That is the conundrum of the day.   

There are pros and cons to both.  With aggregation, you get bigger sample sizes, and at times, it may be easier to work with, but you can also miss important insights.  With disaggregation, you can really dig deeper into populations to understand unique experiences; however, you may have to deal with small sample sizes.  Which way do we go?   We need both aggregation and disaggregation of data.  I worked with an organization whose members were allowed to select from a list of over 500 different race/ethnicity groups.  That information is pulled into dashboards.  The first view of the reports provides the aggregated OMB groups.  However, you can easily access the disaggregated data with a few clicks. This approach allowed us to move work forward with both general and specific insights.  And that is what good data should do.  

Okay, I got a little long-winded in this newsletter.  If you are still here, thank you.  

Much goes into developing a strategy for collecting race, ethnicity, language, sexual orientation, and gender identity data. I would love to see more research on how to apply the frameworks and resources to address these challenges.  Until then, I’ll leave you with one last recommendation.  

Use the data.  Far too many organizations are collecting the information, and it is just sitting on a shelf collecting dust.  Wherever you are on the journey, let the data guide you to ask tough questions, gather insights, and contribute to your decisions.  

Connected Consultants can help you bring together a cross-functional team to develop a REaL and SOGI data strategy. Book your  FREE call here.

Resources

AHA Disparities Toolkit - Staff Training on REaL and SOGI Data

Promoting Interoperability Programs

CMS Sponsored Webinars on Interoperability

Advocating for Data Disaggregation

References

Kauh TJ, Read JG, Scheitler AJ. The Critical Role of Racial/Ethnic Data Disaggregation for Health Equity. Popul Res Policy Rev. 2021;40(1):1-7. doi: 10.1007/s11113-020-09631-6. Epub 2021 Jan 8. PMID: 33437108; PMCID: PMC7791160.

Szarfman A, Levine JG, Tonning JM, Weichold F, Bloom JC, Soreth JM, Geanacopoulos M, Callahan L, Spotnitz M, Ryan Q, Pease-Fye M, Brownstein JS, Ed Hammond W, Reich C, Altman RB. Recommendations for achieving interoperable and shareable medical data in the USA. Commun Med (Lond). 2022 Jul 18;2:86. doi: 10.1038/s43856-022-00148-x. PMID: 35865358; PMCID: PMC9293957.