If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Documentation of sexual orientation (SO) and gender identity (GI) is crucial to identify lesbian, gay, bisexual, and transgender youth and perform meaningful research to improve health disparities in this community. As a result, some electronic medical records (EMRs) have incorporated SO and GI into part of the provider's workflow for documentation. We aimed to evaluate the effect this modification has had on the frequency of SO and GI documentation.
Design, Setting, Participants, Interventions, and Main Outcome Measures
This was a retrospective chart review of patient encounters from an outpatient pediatric and adolescent gynecology clinical practice. The rate of documentation of SO and GI were compared between encounters that took place before the implementation of the EMR modification and those that took place after. Additionally, we examined rates of GI and SO documentation according to visit type and patient race.
A statistically significant increase in the frequency of SO and GI documentation after the EMR modification was detected. The documentation rate of SO increased from 10/73 (13.7%) to 32/73 (45.1%) (P < .01) and GI documentation rate went from 1.4% to 46.5% (P < .01) after the EMR changes were implemented. SO or GI was most commonly documented in social history (90%). There were no differences in documentation on the basis of race or type of encounter.
Including a specific tab for SO and GI in the EMR significantly increased the frequency of SO and GI documentation. Despite this increase, frequency of documentation remained at less than 50%, emphasizing the need for further improvement.
The documentation of an individual's gender identity (GI) and sexual orientation (SO) is often overlooked and possibly never inquired about during a medical encounter. Documentation of this information is crucial for improving research that could assist in providing better inclusive patient-centered care, appropriate patient risk assessment, and data collection to help better understand the health disparities that need to be addressed. The Institute of Medicine noted the need to address the health disparities in the lesbian, gay, bisexual, transgender, and other related communities (LGBT+) because of the increase in medical and social risk factors in the setting of limited access to medical care.
In the 2017 Youth Risk Behavioral Survey (YRBS), the Centers for Disease Control and Prevention reported 47.7% of lesbian, gay, and bisexual (LGB) high school-aged individuals had strongly considered suicide and 23.0% of LGB individuals attempted suicide within the past year.
The 2017 YRBS showed that LGB individuals, compared to their heterosexual peers, are at higher risk for physical dating violence, sexual dating violence, and being forced into sexual activity. LGB individuals were also more likely to be involved with riskier sexual practices such as not using a condom, having more lifetime sexual partners, and increased the risk of certain sexually transmitted infections.
Despite these well documented health disparities, medical care and quality research are lacking in this population for several reasons. The lack of attention to GI and SO in the medical setting is often multifactorial. Transgender youth who do seek medical care often do not return because of anxiety and perceived judgement by medical staff.
Without proper identification and documentation of a patient's SO and GI it is difficult to have sufficient data for research development, which could lead to a better understanding of the health care needs of the LGBT+ community and the development of proper screening guidance.
To improve the documentation of GI and SO, the Promoting Interoperability Guidelines, under the Meaningful Use Stage III, required all certified electronic medical records (EMRs) have a section for GI and SO documentation by January 1, 2018.
As a result of this new requirement, our institution modified the social history section to include tabs specifically addressing GI and SO, which is automatically imported into the provider's note (Fig. 1). This feature was added on September 1, 2017. We performed a retrospective chart review to see if there is an increase in the frequency of documentation of SO and GI data as a result of the EMR modification.
Materials and Methods
This was an institutional review board-approved retrospective cohort study in which documentation rates of SO and GI for Pediatric and Adolescent Gynecology Clinic encounters were compared before and after the introduction of SO and GI tabs in the social history section of the EMR workflow. Encounters for patients between 14 and 18 years old were included. Patients who were nonverbal or had significant developmental delays were excluded from the study. Patient encounters were divided into 2 groups; preimplementation encounters (visits between September 2016 and June 2017) and post implementation (visits between January 2018 and October 2019). These dates were chosen to allow for physician documentation patterns to adjust to the new EMR layout and avoid data contamination.
The sample size was calculated on the basis of a pilot study that collected SO and GI documentation rates from a single month of encounters before the EMR modification. The pilot study showed a documentation rate of 36.0% for SO and 0.0% for GI. We hypothesized that the EMR modification would increase SO documentation by 25.0%. We carried out a formal sample size calculation, which revealed that to attain a power of 80.0% at an alpha level of 0.05, a minimum sample size of 62 encounters would be needed in each group. To ensure sufficient power, we included 75 encounters in each group for data collection of the 221 encounters before and 929 encounters after the EMR implementations. Patient encounters were identified using their unique Financial Identification Number and randomly selected using STATA statistical software (StataCorp LLC).
The provider's note in the EMR was reviewed, in its entirety, for each encounter to identify if SO and GI were documented. The EMR was also used to attain the following data: date of visit, patient's age, race, type of visit (new vs follow-up), GI, and SO (if documented). If GI and SO were documented, the further review examined the area of the chart in which it was documented. It should be noted that providers entered the data in the EMR on the basis of a confidential conversation between the provider and patient to elicit their self-identified SO and GI. Patients 14 years of age and older who are seen in the Pediatric and Adolescent Gynecology clinic are not routinely signed up for the patient portal. Additionally, the patient portal does not contain information from the social history or the history of present illness, however, other providers do have access to this information.
Pearson's χ2 tests were used for group comparisons using STATA statistical software (StataCorp LLC).
There were 1150 patient encounters, 221 before and 929 after EMR implementation, in the Pediatric and Adolescent Gynecology Clinic during the study dates. Of the 150 randomly selected encounters, data from 146 were collected. Two encounters from each group were excluded because of missing provider notes or duplication of Financial Identification Number.
SO was documented in 10 of the 73 (13.7%) preimplementation encounters and 32 of the 73 (45.1%) post implementation encounters (P < .01). GI was documented in 1/73 (1.4%) of the preimplementation encounters and 33/73 (46.5%) of the post implementation encounters (P < .01).
Of the 10 patient encounters that had SO documented in the preimplementation group, 70.0% reported being heterosexual whereas 30.0% reported being homosexual, specifically lesbian. Of the 32 encounters that had SO documented in the post implementation group 94% reported being heterosexual and 6.0% reported being bisexual. The preimplementation group only had 1 documented GI, which was cisgender female. The post implementation group had 33 documented encounters; 94.0% identified as cisgender female whereas 6.0% identified as transgender male.
We further examined if other factors including race or type of visit might influence documentation patterns, but neither factor had a statistically significant effect on documentation rates. The study population was 73/146 (50%) African American, 65/146 (44.5%) Caucasian, 1/146 (0.7%) Hispanic, whereas 4/146 (4.8%) reported “other” or declined to specify. SO documentation was 22/73 (30.1%) in Caucasian, 20/73 (27.4%) in African American, and 0/73 (0%) in Hispanic records (P = .249). SO documentation for new patient vs follow-up visits were 18/71 (25.4%) vs 24/75 (32%) (P = .375), respectively, and GI documentation for visit types were 18/71 (25.4%) and 16/75 (21.3%) (P = .566), respectively.
We reviewed the entire document to find the area within the provider's note where SO and GI were most documented because the SO and GI data are imported from the Social History tab. Of the encounters with documented SO and GI, the information was most often found under Social History 45/50 (90.0%); the remaining 5/50 (10.0%) was found under the History of Present Illness.
The purpose of our study was to determine if the incorporation of SO and GI check boxes into the EMR would increase the frequency of documentation. Our retrospective chart review revealed a documentation increase of more than threefold, going from 10/73 (13.7%) preimplementation to 32/73 (45.1%) post implementation for SO and an increase of 1/73 (1.4%) preimplementation to 33/73 (46.5%) post implementation for GI. These documentation rates support the hypothesis that EMR modifications required under the Meaningful Use Stage III set by the Promoting Interoperability Guidelines increase documentation rates effectively and significantly.
Although there was a significant increase in documentation rates of SO and GI, these rates remained low overall as reported in other studies.
Of importance to note, these elevated documentation rates were in a Pediatric and Adolescent Gynecology Clinic, which might be inclined to document this type of information by nature of the specialty, making these results possibly higher than for other types of clinics. This indicates the need for further documentation improvements of this sensitive and crucial information. LGBT + youth prefer providers to focus on health risks and wellness specifically about them, which starts by identifying their gender and sexuality.
We did not evaluate the level of discussion that occurred on the basis of GI or SO. Studies have shown that members of sexual minority groups face a disproportionate number of health care inequities and often feel uncomfortable in the health care setting.
simply evaluating frequency of documentation does not necessarily reflect meaningful interventions that would influence patient satisfaction or health outcomes.
Although there was no statistically significant difference in the frequency of documentation on the basis of race, there was a trend toward increased documentation of SO among Caucasian patients. With a larger sample size, it is possible that documentation rates would be significantly different. This is important because race is an effect modifier for health care disparities in this high-risk population. For example, a previous study showed that white LGBT + young adults are more likely to be insured.
Interestingly, we did not find a statistically significant difference in follow-up encounters vs new patients visits. Social history should carry over in the workflow, so if it is documented once, all future notes would contain these data. However, we did not find an increase in documentation rates among follow-up patient encounters, which could be linked to the overall low rates of documentation. Carryover of previously entered data might be a determent to the medical record, because SO and GI should be intermittently readdressed. LGBT + youth might not be comfortable honestly disclosing this at their first visit or their self-identification might evolve. Thus, sexuality and gender need to be an ongoing in-depth conversation between patient and provider.
If the EMR automatically populates the previously disclosed SO or GI, a provider might be less likely to readdress.
The study's strengths include the identification of a relevant area for research in the LGBT + population. The study was properly powered with the number of encounters included in data collection exceeding the minimum number needed on the basis of a formal sample size calculation and the study had very limited missing data. The encounters used were randomly selected using statistical software. Finally, the documents reviewed were from multiple different providers. This means that the findings are most likely attributable to the change in the EMR rather than a change in an individual provider's habits.
Although in this study we investigated a novel concept with significant findings, there are limitations. We did not include the reason for the visit in our analysis in documentation rates because of the large amount of variability. There are several reasons patients are seen in this clinic and it is possible that visits specifically for gender-affirming hormones are more likely to have GI or SO documented whereas a patient being seen for an ovarian mass, for example, would be less likely to be asked. We did note the documentation frequency attained from the sample used to determine the power for the study was far higher than that attained in the preimplementation cohort. This might be because the attending physician was responsible for all of the documentation during the month that was selected as the pilot study. Resident physicians became a more active part of the clinic shortly thereafter, so there was a more diverse contribution to physician documentation. It also emphasizes the importance of using a randomized selection of encounters to include. Documentation is an important first step in providing comprehensive care and performing meaningful research that might improve health inequities among LGBT + youth.
Implication and Contributions
Addressing health disparities in gender and sexual minority youth starts with documenting SO and GI in the medical record. An EMR that incorporates SO and GI in the workflow significantly increases the frequency of documentation.
Including SO and GI as distinct entities in the EMRs significantly increases the frequency of documentation of these essential data. Future studies should explore how to further improve trends in documentation necessary to do research that can ultimately improve the health and well-being of LGBT+ youth.
Interviews with patients and providers on transgender and gender nonconforming health data collection in the electronic health record.