The work ahead includes creating an inclusive, respectful culture to promote health-seeking behaviors, identifying, understanding, and eliminating disparities, and developing evidence-based clinical guidelines spanning the lifespan and across the continuum of care. In other posts withing this panel we've discussed the need to create a welcoming environment, accurate and relevant emographic data collection, demographics, encounter element data (e.g. HPI, ROS, PE), and clinical decision support. A successful approach requires ongoing partnership between patients, healthcare professionals, EHR vendors, payors, and the government. All must be aligned to assure optimal population management including quality measures for outcome success metrics. We're gaining momentum, now let's keep moving forward. Here's some helpful background: lgbthealtheducation.org/wp-con...

It's important to keep in mind that the needs of many minoritized communities, including LGBTQ people, change over time. This is true as our evidence base for care evolves, and our understanding of the impact of identity on health. It's critical that all health IT stakeholders keep in mind that it's critical to revisit how we are meeting the needs of all communities. This is not "one and done"!

I lead athena's Patient Safety Team and the Clinical Content team. The latter team manages content assets including terminology, order sets, templates, and screening tools. As a direct result of what I've learned as part of this panel, I'm prioritizing a project to review all of our documentation templates to identify and mitigate any potential gender bias, or unanticipated insensitivity to the LGBTQ perspective. Also, this experience has given me such a wealth of information to share with my R&D and Customer Success colleagues - thanks so much for that!

When I talk to the CMIO at my hospital network, I let that office know that there is research and data to support implementation efforts. We are not in the dark and there is no need to reinvent the wheel.

I, along with other members of our gender care team, have already been working with members of the IT department to make the patient banner in EPIC more user friendly for sexual and gender minorities. This allows for more options that are useful for the clinician. IN addition, I have been part of groups at my state medical association to begin thinking thoughtfully about exactly what we have been discussing this week.

Fortunately, much of this guidance is already published in the literature (by many of the other experts here!). This includes guides on how to ask SO/GI related questions, WPATH's EHR working group as trans-specific recommendations including an anatomy inventory, and multiple LGBTQ institutions offer example language. One possible setback arises that these various resources, while accessible, have multiple audiences. A potential space for progress would be a centralized resource where all interested parties could find relevant information.

An important rule of thumb to use with patient records that indicate a possible trans or gender-expansive identity is to not assume what that identity is or means to that person. Structured data fields are very limited while gender identities and expressions are vast, fluid, and dynamic. As a starting point, a clinician may want to introduce themselves with their name/pronouns to verify the patient's. And, if it's relevant to the interaction (and not out of idle curiosity), check in about language to use for various conditions or body parts. It's a bit like getting patient consent for what is listed in the EHR and how to use that information effectively--with accountability.

I think this is certainly a complex answers. As others have mentioned, this has been looked at in the literature before. I would also suggest working with groups of physicians, NPs, and other users of the EMR to come up with best practices. We have been talking about this for a number of years at our state and national medical associations but it never seems to go anywhere. We are ready and willing to work with vendors and others to design user-friendly systems but those conversations need to be started

My answer here is similar to the one i gave for Matthew's last question. EHRs cannot do the work of effective communication and accountability. They are only a tool to facilitate them. An important rule of thumb to use with patient records that indicate a possible trans or gender-expansive identity is to not assume what that identity is or means to that person. Structured data fields are very limited while gender identities and expressions (like sexual orientations too) are vast, fluid, and dynamic. As a starting point, a clinician may want to introduce themselves with their name/pronouns to verify the patient's. And, if it's relevant to the interaction and not out of idle curiosity, check in about language to use for various conditions or body parts. It's a bit like getting patient consent for what is listed in the EHR and how to use that information effectively--with accountability. And then using that information for clinical decision-making and treatments while also communicating those pathways clearly to the patient.

Physicians and vendors should collaborate to identify and implement the content assets, features, and workflows needed to optimally care for LGBTQ patients. The AMA published a straighforward and easy to implement set of recommendations for Creating an LGBTQ-friendly practice (available at ama-assn.org/delivering-care/p... with links to more in-depth information). This covers everything from providing a welcoming environment to assuring practice standards are met and publicly acknowledging the clinical practice as LGBTQ friendly. Also included are tips that are helpful for vendors creating patient questionnaires and other data collection for demographics, physical assessment, social history, etc..

Community education and reassurance about SOGI are important. The National LGBTQIA+ Health Education Center has patient-facing pamphlets and brochures that can be downloaded at lgbthealtheducation.org and explain why these questions are being asked of all patients, that this information is directly relevant to care, that it is voluntary to provide, and that it will be used confidentially and respectfully, in a manner only directly relevant to the patient's care.

I think that the simplest answer to this question is that it needs to be integrated in a way for clinicians to make meaningful use of that. For example, instead of using a patient's affirmed gender to make clinical decision making, the sex assigned at birth should be used. This means that a doctor will not get alerted to do a required cervical cancer screening on a patient who doesn't have a cervix. Systems that allow doctors to check with organs a patient does and does not have also reduce that burden of EMR prompts that are irrelevant.

It will take a revolution to get there, but imagine if you personally controlled your demographic information through a user-friendly app (that doesn't yet exist) and passed it to your physician or another clinician when it was needed. Their systems would have a way of reading your data and understanding it for purposes of clinical decision support and more. They could store the information that they received in their EMR for legal documentation, billing, and quality/research purposes. But the EMR would no longer be viewed as the source of truth about you.

We can talk all we want about improving EMRs, but the fundamental concept that EMRs are where we should keep our healthcare records is fundamentally flawed because healthcare is no longer done in a single place by a single physician.

Yes, there are exceptions to this, of course. But we won't start addressing them until we turn the "EMR is truth" concept on its head.

BTW, President George Bush had this vision in mind in 2004 when he first funded the Office of the National Coordinator. It soon got hijacked into what we have today, which is a mess.

CDS is an extremely complex issue and there are lots of fine points, but the basic concept is that it should be designed to support the care of the individual patient's needs/situation, not some generalized patient. This can be done by activating filters that can be tweaked by patient and by physician that change the standard CDS to reflect what's needed.

The most effective technology implementation is one in which end users - including their patients - are involved from the very start in the identification of opportunities for improvement, early design, iterative enhancements, and implementation. A key factor is post-implementation monitoring for solution effectiveness. LGBTQ patients should be part of this process to help assess usability from their perspective as patients, including the appropriateness, respectfulness, and relevance of the care they're receiving and the questions that are asked in demographic assessment, screening questions, and throughout the encounter. Outcomes should be assessed, analyzed, and acted upon as with any population, to assure optimal healthcare is provided and undergoes continuous quality improvement.

I would like to see more a more universal approach to SOGI data collection. My health system has started to roll out SOGI data for a subset of the clinics, and there haven't been as much of a focus for the staff/faculties. I think having the transparency in this process is crucial in establishing a new norm, and we need both top-down and bottom-up approaches.

I would like to make sure that staff across the hospital know how to use this data in the EMR, including folks such as clerks, etc. Many people do not know that they are empowered and think that only certain clinics or physicians can change this data when in fact everyone who uses the EMR has access to change this. Staff then need to know that they should reference this banner before just calling out patients names or using incorrect pronouns.

As a small team (read 2 people) the hardest conversations to have are around expertise, data, scale, and usability. When we aren't doctors, it can be difficult to understand what needs are present and what information is necessary in order to design and build a usable, scalable, and interoperable system that works for everyone.

I'd love to see more direct imput from LGBTQ patients about their experiences with care - we've seen some great examples in various questions asked in this AMA panel, but I feel like there's so much more to learn at this time.

The giant push for more SO/GI data is misguided. Data is important, but if the most marginalized don't have access, or consistent access, and don't feel safe disclosing SO/GI when they do access care, then we need to shift our priorities to innovations and community-engaged collaborations to expand access and improve care.

I think the intersectional aspects of LGBTQ health are still growing and developing. I'd love to learn more about how cultures and ethnic minorities intersect with LGBTQ health. For examples, we are beginning to better understand the insidious racism that runs deep within the LGBTQ community, but we still don't fully know how it translates into LGBTQ POC's health outcomes. This is a paradigm that would be dangerous to apply a "one-size-fits-all" approach.

Quality of care for LGBTQIA+ patients is adversely impacted by a lack of research on sexual and gender minority health, which also makes training clinicians difficult since it is then unclear what evidence-based practices and interventions to train clinicians to deliver.

Like any other data field, we need to ensure it's clear *why* we are collecting information. It's too easy to add things, but if they aren't used, aren't actionable, or aren't obviously helpful to the clinician, then they become a burden!

We ask about these things routinely for our adolescents. It only requires about 3 clicks for me to add SO/GI to the chart. I would agree with Dr. Ehrenfeld though, in that we need a purpose for recording it. It shouldn't be in the chart just because we asked and got an answer.

It'd be great to have a streamlined process for SOGI data collection across all EMRs. For EMRs, the SOGI data are still being captured as a free text in a decentralized process; for individuals whose gender pronouns and identities are fluid, they can be a bit cumbersome to update.

It's lunacy to think that we're going to get every EMR retrofitted and every clinician educated and motivated to do the extra work that it takes to record SOGI and other important things. Instead, perhaps let's have the AMA and other professional organizations endorse a standard patient app that has the patient's current demographics and core medical history in a format that their physician's EMR can query (e.g., through FHIR) to download the needed information. If it's cloud-based, the app can be used anywhere by anyone, as long as the patient consents. If all of the clinical professional societies (nursing included), backed this approach and the app was easy to use (that's also important), vendors would build in this capability.

It's a crazy idea, but with the right marketing and support, it's probably far more achievable than the approach we are currently taking.

Clinical documentation n is overly burdensome, we all know that. But quality and reimbursement requirements increase every year, so the vendor focus has to be on shifting the burden of nonclinical work away from the clinician. We actively monitor our network to better understand physician behaviors, for example documentation time including after-hours documentation. As vendors, we have to continuously collect client feedback from across our customer base - not just the enterprise accounts, but also the smaller practices. We have a regular cadence for collecting feedback, synthesizing findings, and advocating to R&D teams - and then holding those teams and their leaders accountable for addressing customer pain points. This is done through surveys throughout the year, acting to improve satisfaction scores, and - most importantly - actively soliciting ideas from across the client base. Submitted ideas are classified into areas of improvement, and then those improvements are identified. this can be done - in 2019, our R&D teams delivered solutions for 60 ideas that were submitted and voted on by customers.

Technologies must be developed with meaningful clinician input throughout the product lifecyle. Not as an afterthought. This is essential to ensuring IT products meet the needs of clinicians!

EHR functionality and clinical workflow must be aligned, supporting the most efficient care model possible to assure staff are working at the top of their scope to coordinate individualized care. The basic tenets of usability (reducing cognitive load, assuring interoperability) are achievable only when end users are directly involved in enhancements. Extensive alpha and beta testing, with iterations based on user feedback, are expensive but worthwhile vendor investments - as is a vendor team of human factors engineers, designers, and user research specialists.

It would be helpful if the EHR Association (within HIMSS) would work together with clinicians and adopt a consensus standard to ensure that the needed information is collected and is consistent across platforms.

+1 to Shannon's answer. To it I would add that almost every vendor has one or more clinicians working for them. These docs and nurses provide their personal guidance about how things should work. But they all have different views of what's important and our professional societies are providing precious little in terms of concrete direction for them

I don't think any of the EHR/HIT vendors are working with the AMA and other professional societies (e.g., the AAP), partly because these professional societies can't seem to get organized to provide meaningful guidance as to what's required or how to set up endorsement requirements that would drive vendors to perform. Until we get to where the professional clinical organizations are integral to the development process, I worry that we'll be slogging through the same swamp forever.

I will echo the voices of others, in that these EHRs were developed for billing, NOT with clinical applicability in mind. Therefore, business needs were the primary driver. It is time to put clinicians back in the driver seat in order to see meaningful change.

Could not agree more. Would you be up for a call to discuss what you find missing in the clinical documentation process?

We're trying to find some physicians to work with closely in developing a solution. Would you be up for a chat to discuss what's needed?