Current Practices in ED Social Determinants Screening and Care Connection: A Literature Review

Samhita N. Basavanhalli (1); Trevor Anderson (2); Jordan Vaughn, MD (1).

Affiliations
1. Louisiana State University Health Sciences Center School of Medicine, New Orleans, Louisiana.
2. Tulane University School of Medicine, New Orleans, Louisiana.

Abstract

Social determinants of health (SDOH) screening in emergency departments (ED) is a promising method to capture and address individualized social needs of a broad patient population, ideally lowering emergency department readmissions while reducing health disparities. With new Joint Commission guidelines requiring social determinants to be addressed and integration of SDOH-related Z-codes into ICD-10 coding, the time is now to implement robust screening and referral programs. This narrative literature review strives to identify best practices prior to the implementation of social determinants screening in the ED of University Medical Center, New Orleans. We investigate current screening tools and their integration with electronic health records, discuss survey formats, detail referral processes, and resource navigation post screening, and describe care connection models from screening to referral. Key conclusions include the identification of the Protocol for Responding to & Assessing Patients’ Assets, Risks & Experiences (PRAPARE) as the ideal screening tool, and that electronic screening tools led to higher levels of social needs reporting compared to paper counterparts. Similar success of written resource referrals and referrals given by a navigator in reducing social risk factors was also identified, highlighting the importance of high-quality, written resource referrals. Lastly, challenges to formation of a successful, integrated screening and referral pathway such as loss to follow-up, even in a transition care coordination model that assists patients throughout levels and types of care, are identified.

INTRODUCTION

Emergency room services play a critical role in public health. According to Ordonez et al. (1), patients with food insecurity, lower education levels, limited access to primary care services, members of racial and ethnic minority groups, and Spanish-speaking patients with limited English proficiency were all correlated with higher ED utilization. Such groups experience systemic barriers in access to care and emergency care transitions. This may lead to greater health disparities, which tend to be strongly associated with increased adverse SDOH (2). SDOH strongly impacts one’s quality of life and life expectancy. According to Alley et al. (3), measurable health outcomes such as mortality and morbidity receive approximately 55% contribution from social, economic, and environmental factors. 

This literature review is being conducted as a review of current practices prior to the introduction of a revamped SDOH screening and referral system in the ED of University Medical Center, New Orleans (UMC). New Orleans and its people are uniquely positioned to reap the benefits of this intervention for many reasons. According to the New Orleans Community Health Improvement Plan (6), only 65% of New Orleanians have a primary care provider, yet chronic conditions are commonplace, with ⅓ of residents suffering from hypertension or hypercholesterolemia and ⅔ of residents considered obese. New Orleans is the second most food insecure city nationally and nearly ¼ of its residents live in poverty, with the city’s average household income being $41,604, over $20,000 under the national average (6). Using the ED to connect New Orleanians to necessary resources such as food banks, housing resources, mental health care, preventative healthcare services, and more will hopefully address some of these issues while lowering ED readmission rates by tackling root causes of admission.

Additionally, with the introduction of a new National Patient Safety Goal by The Joint Commission targeting health equity, hospitals, and other healthcare institutions are more directly incentivized to address social determinants than ever (7). Effective July 1st, 2023, this goal requires institutions to assess patients’ health-related social needs, analyze quality and safety data to identify specific disparities, and develop action plans to improve health equity (7).

Results

Section 1: Discussion of Current Screening Tools 

Henrikson et al. (9) reviewed literature published from 2000 and 2018 to yield 21 unique screening tools for social risk factors in a clinical setting and assessed them on their psychometric and pragmatic characteristics. Tools that are deemed psychometrically strong are able to “accurately and precisely identify social risk domains, characterize their associations with relevant outcomes, and measure changes in risk over time and in response to interventions” (8). Tools that are pragmatically strong were deemed as having favorable pragmatic properties such as ease of administration, low cost, and shorter lengths (9). The top 3 scoring tools for psychometric testing were: Urban Life Stressors Scale, Protocol for Responding to & Assessing Patients' Assets, Risks & Experiences (PRAPARE), and Social Needs Checklist (Henrikson et al., 2019). According to Henrikson et al. (2019), the top 3 scoring screening tools for pragmatic testing were: Survey of Well-Being of Young Children, Safe Environment for Every Kid, and WeCare. 

In a systematic literature review conducted by Chen et al. (5), 4 main SDOH screening tools were focused on. These were the PRAPARE tool, the Accountable Health Communities (AHC) Screening Tool, the Health Leads Screening Tool, and the HealthBegins Upstream Risks Screening Tool. 

According to the literature review, all 4 of the above screening tools cover the 5 key domains outlined by Healthy People 2020, a 10-year program launched by the United States Department of Health and Human Services (HHS) with an objective of improving health through goals like reducing health disparities and reducing preventable disease. These key domains were economic stability, neighborhood and built environment, health and health care, education, and social and community context (5). The review noted that PRAPARE covered the most measures in every domain except health and health care. In this domain, PRAPARE focused on insurance status while the other 3 screening tools focused on needs for assistance, physical activity, diet, and mental health status (5).

Other studies 10 have shown success with utilization of the SDOH screening tool native to EPIC. Benefits included question alignment with other institutional EPIC users, easy access to EPIC population health tools, and quick accessibility of survey responses to team members. 

Utilization of surveys native to mobile apps was also reported (11), including use of HelpSteps, a self-administered screening tools which allows users to choose from 22 different domains of social determinants with accompanying referral options, and simply select their most important, and secondary need domains. This survey was combined with the widely adopted AHC Health-Related screening tool.

Section 2: Discussion on Formats of Surveys: Electronic versus face-to-face screening surveys 

A study conducted by Gottlieb et al. (12), used both electronic and face-to-face screening surveys in a pediatric ED to assess the difference between the two formats. 

The study identified significant differences between the responses from the computer-based surveys and face-to-face interviews, with people being more likely to report social needs items in the computer-based surveys (12).  Respondents reported higher levels of stress related to interpersonal violence (p=0.03) in their homes through computer-based surveys (12). The survey also included higher levels of reported substance use in the home (p=0.05) through computer-based surveys (12). This study is important in demonstrating the significance of the methods of data collection as these methods can affect accuracy and disclosure rates of the patients. A key takeaway from the study is the advantages of using computer-based screening tools, which may be more advantageous as they eliminate feelings of shame or judgment towards patients that may be associated with answering these questions posed directly by a medical provider.

Electronic screening allows surveys to be implemented in a universal fashion, with all patients screened to eliminate non-response bias, as less staffing and administrative resources are needed for administration. This aids in the prevention of certain patients not being screened due to external appearance or demographics. One study (12) found a significant difference in financial insecurity between social screening respondents and non-respondents.

Section 3: Use of Referrals and Resource Navigation 

After SDOH screening through an EHR-integrated tool, results should be used to connect patients to appropriate services. A study conducted by Gottlieb et al. (14) explored the effectiveness of in-person social services navigation assistance in comparison to sharing standardized written information regarding available social resources. The purpose of this study was to investigate methods to make long-term care more feasible and effective in a pediatric urgent care clinic by addressing social risk factors. 

The study randomized patients to receive either written resources or written resources plus in-person assistance. The written resources were prewritten informational handouts that listed local resources from relevant government, hospital, and community social service organizations. The in-person assistance consisted of navigators who also provided other forms of assistance to caregivers such as help with scheduling appointments and completing forms. 

The study found that there were no significant differences between the two groups, but that both groups had significant decreases in reported social risk factors (examples included food insecurity, housing insecurity, and transportation access) as well as improved child and caregiver health. According to Gottleib et al. (14), the results of the study were unexpected as a previous study conducted by the same authors had shown that in-person navigation of resources was significantly more effective. Gottleib et al. (14), discussed that the potential reason for the difference in results could be attributed to the improved quality of information given in the resource sheets. In this study, the navigators incorporated 2 techniques that were recommended by the Agency for Healthcare Research and Quality. These techniques were to include specific contact names at the organizations given and highlight the resources that are most relevant to the social risks identified (14). High-quality written resources may be a sufficient social risk intervention in pediatric populations. 

Applications that automatically generate referrals based on screening responses have also been used, and can be combined with an optional social work consultation.  A study (11) with this approach reported that 14% of the study population reached out to a social support organization.

Community partners, defined as pre-existing organizations that may address specific social determinants of health such as food banks, shelters for the unhoused, or domestic violence prevention programs in addition to programs that focus more broadly on coordinating social needs interventions, are an invaluable resource in executing resource referrals. The strategic utility of these organizations in ensuring high follow-up rates and connection to care cannot be overlooked and strong relationships between healthcare providers and high-use community partners should be cultivated (15). A large academic medical center set up data sharing with an existing community resource directory organization, United Way of Salt Lake City’s 2-1-1. Of the 129 patients with 1 or more stated needs, 73 (56.6%) asked for referral to 2-1-1 and 32 (43.8%) were reached by 2-1-1 within 1 week of emergency department discharge (14). 

A study conducted by Hsieh (16), showed that resource navigators could link patients to primary care providers and other emergency providers. This would allow for continuity and advocacy for the patient’s social concerns. For many high-risk patients, resource navigation may not be sufficient, and establishing ongoing care is a more effective way to intervene in the complex medical issues these patients may be facing. 

Section 4: Putting it All Together: Use of Care Connection Models 

Some papers have described their processes for addressing social needs from start to finish, from initial screening to connection to care or provision of social services.

A systematic review conducted by Yan et al. (17) explored current literature that investigated the process of integrating SDOH or social needs screenings into EHRs and subsequent care connections. They identified three main approaches to identifying and addressing social needs. The simplest approach involved healthcare providers identifying social needs and distributing community resources or referrals as they deemed appropriate. The second approach involved healthcare providers identifying social needs and then, using patient navigators to connect patients to external resources or social services. The third approach, the most complex, involved a transition care coordination model that assisted patients throughout levels and types of care at multiple facilities. Overall, Yan et al. (17) found that while many studies explored the process of integrating SDOH screenings, few studies actually reported health outcome measures. They noted that several studies did report positive impacts on healthcare costs and utilization measures, however, these studies were mixed in their ability to provide conclusive evidence. 

This is congruent with a study conducted by Wallace et al. (18). In this study, the authors evaluated the reach and implementation of integrating SDOH screening and referral to resources in an ED. Between January 2019 to February 2020, ED registration staff screened patients for social needs. They used a 10-item, low-literacy, English-Spanish electronic questionnaire that generated automatic referrals. Wallace et al. (18) found that of the 4608 patients approached, 61% of patients completed the screening questionnaire. Of these patients, 47% indicated a need for one or more social services and 34% of those agreed to be followed up with a resource specialist (18).  Only 20% of those who agreed to be followed up with were reached out to by outreach specialists for referrals. Only 7% of patients completed the process from screenings to referrals. This overall low completion rate should be considered when implementing referral processes. The article then explored the challenges that arose during this process such as patient stigmatization and staff reluctance. Detailed evaluation of the process determined that patients desired a better understanding of their needs and had felt concerns regarding privacy and being stigmatized from the screening staff. The screening staff expressed discomfort and that they were questioning the usefulness of screening for social needs.

Some authors have described utilizing automation to increase efficiency from screening to referral. An article by Rogers et al. (19) described a custom screening tool they built into Epic EHRs. The screening tool used was reflective of the AHC screening tool, mentioned in Section 1. Rogers et al. (19) customized the tool by integrating it with a Community Resource Network Management Software-as-a-Service (CRNM SaaS). The steps of the process began with the AHC screening tool. Then, these responses were recorded in the patient’s EHR and transmitted from EHR to CRNM SaaS platform. The CRNM software reviews the screening results and automatically generates a customized community resource sheet (CRS)  that can be given to the patient with their After Visit Summary (AVS). This tailored CRS includes community service providers (CSPs) in the patient’s ZIP code or nearest ZIP code that could assist with each SDOH identified in the AHC tool. The strengths of this program were the reduced burden on healthcare staff. Additionally, providing patients with a customized CRS can help account for language or literacy barriers that may prevent the patient from using the information listed on the sheet.

Section 5: Our Proposed Intervention as Informed by the Literature

As described in the table above, we recommend implementing a program similar to the one described by Rogers et al.19 Based on the demographics in New Orleans and the patient population at UMC, we propose utilizing a social determinants screening tool built into EPIC as well as the integration of a program called FindHelp into Epic. FindHelp uses a unique platform to connect people to local resources and programs. We have outlined a flow chart of our proposed intervention in Figure 1. 

The recommended process will be as follows. When a patient arrives at the UMC ED, they will receive a link to MyChart. MyChart is a secure location that stores a patient’s health information including medications, medical bills, test results, and appointments. Through MyChart, the patient will be able to complete a SDOH screening survey (Figure 2).

Figure 1. Flow chart through patient arrival at ED to resource referral upon discharge.

Figure 2. SDOH Screening Survey to be implemented

Physical Activity 

• On average, how many days per week do you engage in moderate to strenuous exercise (like a brisk walk)?

• 0 days

• 1 day

• 2 days

• 3 days

• 4 days

• 5 days

• 6 days

• 7 days

• Patient refused 

• On average, how many minutes do you engage in exercise at this level?

• 0 min

• 10 min

• 20 mins

• 30 mins

• 40 mins

• 50 mins

• 60 mins

• 70 mins

• 80 mins

• 100 min

• 120 min 

• 130 min

• 140 min 

• 150+ mins

Financial Resource Strain

• How hard is it for you to pay for the very basics like food, housing, medical care, and heating?

• Very hard

• Somewhat hard

• Not very hard

• Not hard at all

• Patient refused 

Housing Stability

• In the last 12 month, was there a time when you were not able to pay the mortgage or rent on time?

• Yes

• No 

• Patient refused 

• In the last 12 months how many places have you lived (open response). 

• In the last 12 months, was there a time when you did not have a steady place to sleep or slept in a shelter (including now)?

• Yes

• No

• Patient refused

Transportation Needs 

• In the past 12 months, has lack of transportation kept you from medical appointments or from getting medications? 

• Yes

• No 

• Patient refused

• In the past 12 months has lack of transportation kept you from meetings, work, or from getting things needed for daily living?

• Yes 

• No 

• Patient refused

Food insecurity 

• In the past 12 months, you worried that your food would run out before you got the money to buy more?

• Never true

• Sometimes true 

• Often true

• Patient refused

• Within the past 12 months, the food you bought just didn't last and you didn't have money to get more?

• Never true

• Sometimes true 

• Often true

• Patient refused

Stress

• Do you feel stress- tense, restless, nervous, or anxious or unable to sleep at night because your mind is troubled all the time- these days?

• Not at all

• Only a little

• To some extent

• Rather much

• Very much

• Patient refused

Social Connections

• In a typical week, how many times do you talk on the phone with family, friends or neighbors?

• Never

• Once a week

• Twice a week

• Three times a week

• More than 3 times a week

• Patient refused

• How often do you get together with friends or relatives?

• Never

• Once a week

• Twice a week

• Three times a week

• More than 3 times a week

• Patient refused

• How often do you attend church or religious services?

• Never

• 1 to 4 times per year

• More than 4 times per year

• Patient refused

• Do you belong to any clubs or organizations such as church groups, unions, fraternal or athletic groups, or school groups?

• Yes

• No

• Patient refused

• How often do you attend meetings of the clubs or organizations you belong to?

• Never

• 1 to 4 times per year

• More than 4 times per year

• Patient refused

• Are you married, widowed, divorced, separated, never married or living with a partner?

• Married

• Widowed

• Divorced

• Separated

• Never Married

• Living with partner

• Patient refused

Intimate Partner Violence 

• Within the last year, have you been afraid of your partner or ex-partner?

• Yes 

• No 

• Patient refused

• Within the last year, have you been humiliated or emotionally abused in other ways by your partner or ex-partner?

• Yes

• No 

• Patient refused

• Within the last year, have you been raped or forced to have any kind of sexual activity by your partner or ex-partner? 

• Yes

• No

• Patient refused

Alcohol Use

• How often do you have a drink containing alcohol?

• Never

• Monthly or less

• 2-4 times a month

• 2-3 times a week

• 4 or more times a week

• Patient refused

• How many drinks containing alcohol do you have on a typical day when you are drinking?

• Patient does not drink

• 1 or 2

• 3 or 4 

• 5 or 6 

• 7/9

• 10+

• Patient refused

• How often do you have six or more drinks on one occasion?

• Never

• Less than monthly

• Monthly

• Weekly

• Daily or almost daily

• Patient refused

Utilities

• In the past 12 months has the electric, oil, or water company threatened to shut off services in your home?

• Yes

• No

• Already shut off

• Patient refused

The patient will then be sent to triage where the triage nurse will verify survey’s completion, or answer any questions to facilitate its completion. The survey tool will be added to the EPIC toolbar for easy accessibility to staff. If the survey is not completed pre-triage, or during triage, it can also be completed afterward while waiting for care. Once the patient has been treated and is ready to be discharged, a curated list of community organizations and referrals will be sent home with the patient based on the SDOH FindHelp identified from the survey. A case manager will also be assigned to the patient to follow up with them and provide any additional assistance. 

Potential weaknesses of this program include its reliance on patients having a mobile device to access the internet. The MyChart link will be sent to a patient’s phone either through text or email. If a patient does not have a phone, we hope to have ED iPads that can be used to fill out the screening surveys. Another weakness is that this program may be difficult for patients with low literacy levels, low proficiency in English, or disabilities. Providing iPads with accessibility features could help combat this issue but would require staff to be available to assist these patients. 

Discussion

SDOH screening is an important and growing objective in social Emergency Medicine. Beyond its importance in reducing hospital readmission rates by addressing root cause of disease or ED presentation, screening efforts address the new National Patient Safety Goal by The Joint Commission, and have recently been integrated in ICD-10 coding. 

Z codes are a separate set of ICD-10 codes that can be used to document patients’ SDOH.22 They include a wide range of issues such as education & literacy, employment, housing status, access to food, access to safe drinking water, occupational hazards, and more.22 The Centers for Medicare & Medicaid Services’ Office of Minority Health23 released a report in June 2023 that maps out steps to effectively using Z codes. The steps they recommend are the following: (1) collect SDOH data, (2) document the SDOH data in the patient’s record, (3) map SDOH data to Z codes, (4) use SDOH Z code data, (5) report SDOH Z code data findings. There are several benefits to collecting this information such as helping the hospital and healthcare staff identify the most commonly used Z codes. Identifying top Z codes can help focus referrals and resources in those specific areas and help to efficiently reduce SDOH. 

There is also ample room for future research in this arena, such as an evaluation of follow-up rates with referral services to determine if the resources are being used and to what extent. Future research can also be done to explore SDOH screening implementation in settings other than critical/urgent care settings. 

The role and efficacy of healthcare systems in implementing their own social needs interventions that do not require support from community organizations is another area the literature is lacking. Social determinants screening data can potentially be used to tailor interventions relevant to hospitals’ unique patient populations. For example, a large proportion of patients indicating food insecurity during screening may indicate that a hospital-run food pantry would be beneficial. By implementing interventions without the reliance on partner organizations post-referral, healthcare entities may be better able to follow-up with patients and connect them to resources in a timely manner post-screening. 

Following these recommendations for the SDOH screening process in the ED of UMC, robust collaboration, feedback, and training will be needed to ensure this new process is streamlined and effective for all stakeholders. After its complete rollout, data analysis and quality improvement initiatives will be necessary to ensure completion rates are as high as possible and that patients are being connected to needed resources in a timely and efficacious manner.

References

1. Ordonez E, Dowdell K, Navejar NM, Dongarwar D, Itani A, Salihu HM. An Assessment of the Social Determinants of Health in an Urban Emergency Department. West J Emerg Med. 2021;22(4):890-897. Published 2021 Jul 15. doi:10.5811/westjem.2021.4.50476

2. Khidir H, Salhi R, Sabbatini AK, et al. A Quality Framework to Address Racial and Ethnic Disparities in Emergency Department Care. Ann Emerg Med. 2023;81(1):47-56. doi:10.1016/j.annemergmed.2022.08.010

3. Alley DE, Asomugha CN, Conway PH, Sanghavi DM. Accountable Health Communities–Addressing Social Needs through Medicare and Medicaid. N Engl J Med. 2016;374((1)):p. 8–11. doi: 10.1056/NEJMp1512532.

4. Walter LA, Schoenfeld EM, Smith CH, et al. Emergency department–based interventions affecting social determinants of health in the United States: A scoping review. Academic Emergency Medicine. Published online February 2, 2021. doi:https://doi.org/10.1111/acem.14201

5. Chen M, Tan X, Padman R. Social determinants of health in electronic health records and their impact on analysis and risk prediction: A systematic review. J Am Med Inform Assoc. 2020;27(11):1764-1773. doi:10.1093/jamia/ocaa143

6. New Orleans Health Department. New Orleans Community Health Improvement Plan. New Orleans Health Department; 2022. https://nola.gov/getattachment/Health/Community-Health-Improvement/Reports/NOHD_New-Orleans-CHIP-2022-2025_FINAL.pdf/?lang=en-US

7. The Joint Commission. R3 Report Issue 38: National Patient Safety Goal to Improve Health Care Equity. The Joint Commission R3 Report. Published December 20, 2022. https://www.jointcommission.org/-/media/tjc/documents/standards/r3-reports/r3-report_npsg_16.pdf

8. Paré G, Kitsiou S. Handbook of EHealth Evaluation: An Evidence-Based Approach, Chapter 9: Methods for Literature Reviews. University of Victoria; 2017. https://www.ncbi.nlm.nih.gov/books/NBK481583/#

9. Henrikson NB, Blasi PR, Dorsey CN, et al. Psychometric and Pragmatic Properties of Social Risk Screening Tools: A Systematic Review. Am J Prev Med. 2019;57(6 Suppl 1):S13-S24. doi:10.1016/j.amepre.2019.07.012 

10. Peretz P, Shapiro A, Santos L, et al. Social Determinants of Health Screening and Management: Lessons at a Large, Urban Academic Health System. Jt Comm J Qual Patient Saf. 2023;49(6-7):328-332. doi:10.1016/j.jcjq.2023.04.002

11. Kanak MM, Fleegler EW, Chang L, et al. Mobile Social Screening and Referral Intervention in a Pediatric Emergency Department. Acad Pediatr. 2023;23(1):93-101. doi:10.1016/j.acap.2022.08.011

12. Gottlieb L, Hessler D, Long D, Amaya A, Adler N. A randomized trial on screening for social determinants of health: the iScreen study. Pediatrics. 2014;134(6):e1611-e1618. doi:10.1542/peds.2014-1439

13. Vest JR, Mazurenko O. Non-response Bias in Social Risk Factor Screening Among Adult Emergency Department Patients. J Med Syst. 2023;47(1):78. Published 2023 Jul 22. doi:10.1007/s10916-023-01975-8

14. Gottlieb LM, Adler NE, Wing H, et al. Effects of In-Person Assistance vs Personalized Written Resources About Social Services on Household Social Risks and Child and Caregiver Health: A Randomized Clinical Trial. JAMA Netw Open. 2020;3(3):e200701. Published 2020 Mar 2. doi:10.1001/jamanetworkopen.2020.0701

15. Wallace AS, Luther B, Guo JW, Wang CY, Sisler S, Wong B. Implementing a Social Determinants Screening and Referral Infrastructure During Routine Emergency Department Visits, Utah, 2017-2018. Prev Chronic Dis. 2020;17:E45. Published 2020 Jun 18. doi:10.5888/pcd17.190339

16. Hsieh D. Achieving the Quadruple Aim: Treating Patients as People by Screening for and Addressing the Social Determinants of Health. Ann Emerg Med. 2019;74(5S):S19-S24. doi:10.1016/j.annemergmed.2019.08.436

17. Yan AF, Chen Z, Wang Y, et al. Effectiveness of Social Needs Screening and Interventions in Clinical Settings on Utilization, Cost, and Clinical Outcomes: A Systematic Review. Health Equity. 2022;6(1):454-475. Published 2022 Jun 24. doi:10.1089/heq.2022.0010

18. Wallace AS, Luther BL, Sisler SM, Wong B, Guo JW. Integrating social determinants of health screening and referral during routine emergency department care: evaluation of reach and implementation challenges. Implement Sci Commun. 2021;2(1):114. Published 2021 Oct 7. doi:10.1186/s43058-021-00212-y

19.  Rogers CK, Parulekar M, Malik F, Torres CA. A Local Perspective into Electronic Health Record Design, Integration, and Implementation of Screening and Referral for Social Determinants of Health. Perspect Health Inf Manag. 2022;19(Spring):1g. Published 2022 Mar 15.

20. The Joint Commission. Assess Health-Related Social Needs. The Joint Commission Accreditation Resource Center . Published 2023. https://www.jointcommission.org/our-priorities/health-care-equity/accreditation-resource-center/assess-health-related-social-needs/#t=_StrategiesTab&sort=%40created%20descending

21. Truong HP, Luke AA, Hammond G, Wadhera RK, Reidhead M, Joynt Maddox KE. Utilization of Social Determinants of Health ICD-10 Z-Codes Among Hospitalized Patients in the United States, 2016-2017. Med Care. 2020;58(12):1037-1043. doi:10.1097/MLR.0000000000001418

22. Maksut J, Hodge C, Razmi A, Khau M. Utilization of Z Codes for Social Determinants of Health among Medicare Fee-For-Service Beneficiaries, 2019. Centers for Medicare and Medicaid Services; 2021. https://www.cms.gov/files/document/z-codes-data-highlight.pdf

23. Using Z Codes: The Social Determinants of Health (SDOH) Data Journey to Better Outcomes. Centers for Medicare & Medicaid Services; 2023. https://www.cms.gov/files/document/zcodes-infographic.pdf

The Expansion of Private Equity into Ophthalmology

Muhammad Awan, BS (1); Ankit Shah, MD (1).

Affiliations
1. Alabama College of Osteopathic Medicine, Dothan, Alabama.

Abstract 

Throughout the last few decades, private equity has expanded into the field of ophthalmology at an  alarming rate. Acquisitions of private practices by firms have increased with the intention of maximizing  profits and patient volume in order to target the growing demand for medical services and address a  fragmented healthcare practice system [2]. Private equity investment has proven to greatly benefit private  equity firms and senior ophthalmologists of well-established practices alike. However, this trend of  private equity acquisition has received much criticism despite its seemingly positive short-term outlook.  In this commentary, we discuss the potential setbacks of private equity advancement in ophthalmology, a  discussion which may also be applied to other specialties experiencing a similar predicament. Loss of  physician autonomy, declining physician income for junior ophthalmologists, and the priority of  profitability over patient care are a few of the significant concerns highlighted in this commentary  regarding private equity. Furthermore, we consider the ramifications of private equity investment on  incoming ophthalmologists who are entering an uncertain job marketplace and may struggle to locate  stable practice opportunities [8]. This commentary concludes with the evaluation of private equity  advancement through the lens of a medical student and an ophthalmologist, as well as the call for medical  students and trainees to educate themselves on the matter and promote further research on the long-term  consequences of this trend of private equity investment. 

Commentary 

Private equity has made significant advancements into the field of ophthalmology throughout the past few  decades, most notably with a trend of increasing private equity acquisitions of private practices. For  medical students, residents, and trainees in the early part of their career, this is an evolving dynamic that  requires attention due to its paramount effect on the private practice landscape. This commentary attempts  to further educate students, residents, and early career physicians on the impact of private equity’s  encroachment on physician practices, specifically in the ophthalmology space. Most of these findings  may relate to other specialties, such as dermatology and radiology. 

Private equity firms seek to maximize profit through acquisitions of private practices, with the aim of  increasing the value of purchased ophthalmology practices and thereupon selling the practice to another  firm that shall do the same. These firms are thus understandably drawn toward more profitable private  practices that are highly likely to grow financially [1]. Multiple ophthalmology practices may also be  consolidated into larger groups under the same private equity firm to increase profitability, with the  providers of these practices becoming employees under these firms. 

This expansion of private equity into ophthalmology has been justified with multiple motives, including a  growing demand for medical services from an aging United States population and targeting “inherent  inefficiencies present in a relatively fragmented practice environment” [1, 2]. Private equity investment of  ophthalmology practices has demonstrated short-term financial success for both senior ophthalmologist  providers and the private equity firm itself due to increased profits, increased patient volume, and  improved payer mix [3]. 

Nonetheless, there remains much skepticism regarding the true success of private equity acquisitions of  private practices in the long-term perspective. Despite the beneficial short-term patterns mentioned above, other studies have also demonstrated decreased physician autonomy and physician salaries following  acquisition. Due to the uncertainty regarding the fate of junior and incoming ophthalmology colleagues, private equity encroachment on ophthalmology practices raises a serious concern for the future of  ophthalmology and patient care.  

The decrease in physician autonomy results from ophthalmology providers handing clinical operations authority to the private equity firm. With the acquisition of practices by private equity firms,  ophthalmologists must then operate under certain policies imposed by the larger business practice model.  Subsequently, these firms then restructure the ophthalmology practice to establish an organization that  primarily seeks financial success over patient satisfaction and optimal patient care [4, 5]. The decline in  physician reimbursements is also influenced by the focus of private equity on financial efficiency. Private  equity firms maximize profitability of ophthalmology practices by reducing costs of care, implementing  structural changes, and increasing volume of service, among other methods. However, these strategies  together may necessitate tighter control over expenses and ophthalmologist income.  

Furthermore, multiple studies have suggested that junior and future ophthalmologists are “less likely to  succeed financially compared with their contemporaries” [6] due to private equity investment. Junior  ophthalmologists typically join large physician-owned practices with the goal of eventually becoming a  partner, a process that requires buying into the practice after a pre-determined employment period. This  process often takes several years for both junior and senior ophthalmologists. Both sides have to decide if  the “marriage” is the right fit. For example, the junior ophthalmologist needs to determine if the practice  model, philosophy, the patient clinical and surgical volume, and the practice environment are mutually  beneficial. The senior ophthalmologist must decide whether the junior colleague’s personality, clinical  and surgical expertise, and work ethic align with the practices. Once both sides have mutually decided to  pursue a buy-in, a pre-determined price is then paid by the junior ophthalmologist over a period of time,  after which subsequent profits of the practice are then appropriately distributed to all of the partners  including the new junior ophthalmologist.  

Unfortunately, the junior ophthalmologist is highly susceptible to the risk of a private equity firm buying  out the practice before the physician can secure their buy-in. If a firm does approach the practice while  the junior physician’s buy-in is not yet complete, this ophthalmologist is not considered a partner and  hence is excluded from the negotiations with the private equity firm. The junior ophthalmologist is not  included in the payout agreed upon by the partners. Consequently, the junior colleague loses their ability  to buy the practice or become a partner, rendering their efforts over the last few years entirely useless.  Additionally, once the buyout has been completed, the junior ophthalmologist loses all autonomy and will  likely be asked to streamline clinical operations based on the private equity firm’s recommendations. 

From the perspective of a practicing ophthalmologist, private equity encroachment on ophthalmology  practices has been met with mostly negative reviews. The three biggest issues include loss of autonomy  and independence, declining reimbursement, and a practice pattern built on profitability and not  necessarily patient outcomes. Most private equities, once they purchase a practice, require senior  providers who were paid out to remain on board for several years to help the transition to a private equity owned practice. In this time frame, there is significant clinical reorganization, where metrics are  implemented to increase patient volume (clinical and surgical), optimize billing and collections, and  reduce costs. This often results in adding more patient appointment times, reducing visit durations,  reducing salaries, and increasing thresholds for bonus compensation structures. Additionally, product  replacement and medical equipment purchasing are reduced to offset costs. Some groups may align  compensation with patient satisfaction scores which often does not correlate with each other. The  difficulty ultimately comes from the fact that the interests of the private equity company usually do not 

align with those of the physician. Furthermore, solo and independent ophthalmology practices are left to  compete against large private equity firm-owned ophthalmology practices who may more resources and  finances to advertise, recruit, and retain patients, staff and physicians. This dichotomy becomes even  greater as more practices become absorbed by private equity firms. 

The increased presence of private equity in ophthalmology is a movement that also cannot be ignored and  should be understood by all medical students intrigued by the field. Through a medical student and future  ophthalmologist standpoint, the expansion of private equity into ophthalmology raises significant concern  due to the lack of evidence regarding its long-term success. This trend also challenges the comfort of  

physician autonomy and personalized patient care that have always been synonymous with the concept of  private practices. Although private equity acquisition is associated with increased patient volume, the  reality is that the focus of profit over patient care pressures ophthalmologists to see more patients in less  time, resembling an unfavorable move away from personalized care. Less time for patients equates to  poorer patient care, fragile physician-patient relationships, and greater patient dissatisfaction. 

Patient care is the foundation of medicine, and making patient care secondary to profitability poses great  danger to both the patient and the integrity of medical practices. Students are inspired by fields such as  ophthalmology that entail established and prolonged relations with their patients, but the increasing  involvement of private equity in ophthalmology may deter many trainees who are seeking these fond  relationships. Furthermore, the anticipated difficulty of becoming a partner at physician-owned practices and an indeterminate job marketplace present much uncertainty in the future of incoming  ophthalmologists [7, 8].  

With private equity acquisitions on the rise in ophthalmology, it is imperative that medical students,  residents, and junior ophthalmologists are knowledgeable on the impact of private equity on the future of  this specialty. Education regarding private equity and practice options should be widely provided to  trainees pursuing ophthalmology and other specialties that are being affected by the expansion of private  equity investment. Although several highly accountable publications have exposed the trends of private  equity encroachment on medical practices, additional research investigating the long-term implications of  private equity acquisition and its impact on medicine must be done.

References 

1. Sridhar, Jayanth, and Christina Y Weng. “Editorial: Private equity investment and ophthalmology: why  the discussion matters.” Current opinion in ophthalmology vol. 33,5 (2022): 339-341.  doi:10.1097/ICU.0000000000000871 https://pubmed.ncbi.nlm.nih.gov/35916563/ 

2. Del Piero, Juliet et al. “Driving forces and current trends in private equity acquisitions within  ophthalmology.” Current opinion in ophthalmology vol. 33,5 (2022): 347-351.  doi:10.1097/ICU.0000000000000880 https://pubmed.ncbi.nlm.nih.gov/35838270/ 

3. Brill, Daniel et al. “Private equity in ophthalmology: lessons from other specialties.” Current opinion  in ophthalmology vol. 33,5 (2022): 352-361. doi:10.1097/ICU.0000000000000876 https://pubmed.ncbi.nlm.nih.gov/35916564/ 

4. Christopher Kent, Senior Editor. “Update: Private Equity in Ophthalmology.” Review of  Ophthalmology, 10 May 2022. www.reviewofophthalmology.com/article/update-private-equity-in-ophthalmology.  

5. Scheffler, Richard, et al. Soaring Private Equity Investment in the Healthcare Sector, 18 May 2021.  https://bph-storage.s3.us-west-1.amazonaws.com/wp-content/uploads/2021/05/Private-Equity-I Healthcare-Report-FINAL.pdf 

6. Shah, Chirag P, and Jeremy D Wolfe. “How private equity achieves return on investment in  ophthalmology.” Current opinion in ophthalmology vol. 33,5 (2022): 362-367.  doi:10.1097/ICU.0000000000000879 https://pubmed.ncbi.nlm.nih.gov/35819901/ 

7. Portney, David S et al. “Trainee Perspectives of Private Equity's Impact on Ophthalmology.” Journal  of academic ophthalmology (2017) vol. 15,1 e56-e61. 9 Feb. 2023, doi:10.1055/s-0043-1761289 https://pubmed.ncbi.nlm.nih.gov/38737149/ 

8. Patel, Shriji et al. “Implications of the presence of private equity in ophthalmology: an academic  perspective.” Current opinion in ophthalmology vol. 33,5 (2022): 377-380.  doi:10.1097/ICU.0000000000000856 https://pubmed.ncbi.nlm.nih.gov/35819904/

Assessing the Need to Educate Prehospital Providers on the Sex Differences in the Clinical Presentation of Acute MI

Anna Slebonick (1); Kristen Ryczak (1).

Affiliations
Drexel University College of Medicine, Philadelphia, Pennsylvania.

Contact
Anna Slebonick: ams3283@drexel.edu; Kristen Ryczak, kmr443@drexel.edu

Abstract

Medical research has historically underrepresented females. As part of a movement to better represent women in research, U.S. Congress passed the National Institute of Health Revitalization Act of 1993, which mandated the inclusion of women and minority groups in clinical research.1 With the increase in female inclusion, sex differences in the clinical presentations of diseases and responses to medication have emerged. Previous research has found differences in how women and men might present with acute myocardial infarction (MI), with women more often experiencing signs and symptoms that are labeled as ‘atypical’ or ‘nontraditional.’2-8 However, this information has not been robustly incorporated into the prehospital curriculum standards, including education for emergency medical technicians (EMTs) or paramedics.9 This paper investigates whether there is a need for the prehospital curriculum to discuss the lesser-known signs and symptoms of acute MI and the differences in disease presentation between men and women. First, some differences between women’s and men’s physiology and pathophysiology are explored. Next, the clinical presentation of acute MI is reviewed, including the research supporting the sex differences in clinical presentation. Additionally, the disparities in EMS quality of care will be discussed, as research has shown that women receive lower quality of care in the prehospital setting.10 Social factors that can affect delays to treatment will also be discussed. Since there is substantial evidence that health outcomes of acute MI differ between men and women, EMS education should include and emphasize these topics.

Methodology

This integrative review identified appropriate articles through a search on PubMed with the keywords “acute MI,” “sex differences,” “gender differences,” “sex and gender differences in the clinical presentation,” “pathophysiology,” “anatomy,” “differences in acute MI clinical presentation of transgender individuals,” “EMS recognition of acute MI,” and “time to treatment and outcome.” The PubMed search did not include “chromosomal anomalies,” as EMS provider educational standards do not include knowledge of genetic diseases and chromosomal anomalies.9 The search included articles published after 2000. These articles were then assessed and considered relevant if they addressed the following questions:

• Are there sex differences in the clinical presentation of acute MI? Which sexes are discussed in each study?

• Are there anatomical, pathophysiological, or hormonal differences in women and men that affect the development of acute MI? If there are differences, do they affect the signs and symptoms that patients experience?

• Is there an association between time to treatment for acute MI and functional outcomes or mortality? If so, what is the significance in relation to EMS care?

• How accurate is EMS provider recognition of acute MI?

• Is there a difference in the quality of care between EMS providers caring for male and female patients?

Sex and Gender

A person’s chromosomal makeup of XX (female) or XY (male) defines one’s sex. Other chromosomal anomalies exist, such as XXY (Klinefelter’s Syndrome) or X (Turner’s Syndrome). Gender is defined as how a person chooses to identify and express themselves. Gender exists as a spectrum and includes categories such as man, woman, transgender man or woman, or nonbinary, as well as others. All the reviewed studies placed patients into binary categories of “male” or “female.” None of the reviewed studies included individuals with chromosomal anomalies or those who identify as non-cisgender. Similar to the reviewed studies, this paper will focus on the differences between biological males and females. The conclusion of this paper emphasizes the importance of and the need for acute MI research to encompass a wider spectrum of gender inclusion.

Body

Acute Myocardial Infarction

Most people experiencing an acute MI will present with the well-known signs and symptoms that are taught to healthcare professionals: chest pain, pressure, tightness or discomfort, and diaphoresis. However, numerous studies have found that, compared to men, women more frequently present with additional symptoms that are not as well-taught. Women more commonly present with nausea, vomiting, stomach pain, indigestion, heart palpitations, and dyspnea (2,3). While men more often report pain or discomfort in their left shoulder, women more commonly experience pain or discomfort in their jaw, neck, arms, or between their shoulder blades (2,3). Women are also more likely to have fatigue as their only symptom (3). Although chest pain often accompanies these additional symptoms in both men and women, it may be absent in either sex. However, certain female demographics may disproportionately not experience chest pain at all. Lichtman et al. found that, compared to men, women aged 18 to 55 more often present without chest pain during an ST-elevated acute MI (2). Since EMS providers commonly care for patients experiencing an acute MI, they should be educated about the unfamiliar symptoms that may present with or without chest pain and also be informed that women more commonly experience these less frequently emphasized symptoms. 

The differences between men's and women’s vascular physiology and the development of vascular disease may partly account for differences in clinical presentations, though this relationship has yet to be definitively determined. First, women’s coronary arteries are narrower than men’s, and women have a higher baseline myocardial blood flow (11). Haider et al. suggest that these two factors could contribute to a higher degree of endothelial shear stress, potentially helping to prevent plaque accumulation in women’s coronary arteries (11). The pathophysiology of cardiovascular disease has also been shown to differ between sexes. A study from 2021 found that plaque erosion is responsible for nearly one-third of acute coronary syndrome (ACS) incidents, including unstable angina and acute myocardial infarctions (12). Plaque erosion occurs when the top endothelial layer of a plaque is lifted, and platelets and fibrin deposit there and form a white thrombus (12). Plaque rupture is responsible for the remaining ACS incidents, which can occur when an atherosclerotic plaque partially lifts off the luminal wall of a blood vessel, and fibrin and red blood cells deposit and form a red thrombus at the site of the break (12).  Studies have found that plaque rupture is more often experienced by men, and plaque erosion is more commonly experienced by women (11,13).  However, one study found that as women’s age increased, the prevalence of ACS due to plaque rupture also increased (13). While there is clear evidence that men’s and women’s vascular physiology and disease development differ, further research is needed to definitively determine whether differing patterns of plaque buildup and thrombosis directly contribute to differences in men’s and women’s clinical presentation of acute MI. 

In addition to vascular pathophysiology, differences in men’s and women’s hormonal physiology also affect cardiovascular health. Premenopausal women have more circulating estrogen compared to men or postmenopausal women (13). Estrogen has an anti-inflammatory effect on blood vessels and favors low vascular resistance. Studies have shown that premenopausal women have a lower incidence of cardiovascular disease compared to postmenopausal women and men (11,14). Men have more testosterone than women, but research has found conflicting results regarding the overall effect of testosterone on cardiovascular health (15). More research is needed to determine whether there is a direct link between different hormonal profiles and disease presentation. There are studies examining the effects of estrogenic treatments such as oral contraceptive pills and post-menopausal hormone therapy on acute MI risk; however, this information lies outside the scope of EMS education. An in-depth discussion of hormones requires a high-level understanding of endocrinology, which is generally not included in EMS provider education. 

Myocardial infarction pathophysiology also differs between women and men. ACS syndromes are classified as Type 1 and Type 2, with ACS Type 1 accounting for roughly 90% of cases and ACS Type 2 accounting for the remaining 10%. ACS Type 1 is defined as an acute atherothrombotic event. A greater proportion of women experience ACS Type 2, which occurs when an infarction is caused by inadequate myocardial oxygenation without injury to the coronary arteries.3 Inadequate oxygenation can occur due to an increase in oxygen demand and/or a decrease in supply. Tachycardia or hypertension can increase myocardial oxygen demand, and hypoxemia, anemia, and hypotension decrease oxygen supply. Often, ACS Type 2 occurs in the presence of more than one of these conditions (16). For instance, ACS Type 2 is associated with operations, sepsis, arrhythmia, and anemia, which all could reasonably lead to an increase in myocardial oxygen demand and/or a decrease in oxygen supply (17). There is limited information discussing sex differences underlying the mechanism of ACS Type 2. Nevertheless, educating prehospital providers about the potential sex differences in heart disease pathophysiology could facilitate their understanding of the differences in men's and women’s clinical presentation. 

The patient’s self-reported history can significantly impact the prehospital provider’s understanding of the patient’s condition. Interestingly, Lichtman et al. found that women were more likely to perceive their symptoms as anxiety or stress, whereas men were more likely to perceive the symptoms as muscle pain (2). Therefore, a woman with chest pain and lesser-known acute MI symptoms may tell the prehospital provider that her symptoms are anxiety-related. If the EMS believes the patient is experiencing acute anxiety and does not evaluate for an acute MI, several unfortunate scenarios could result in worse outcomes for the patient. Theoretically, EMS may fail to perform an EKG, assign the patient as low priority, which places the patient at risk of seriously deteriorating as they wait for an emergency department (ED) assessment. Or worse, EMS may recommend that the patient try to remain calm at home and call 911 again if they still have symptoms later. In either case, the patient would experience a delay in treatment and potentially worse outcomes. Another factor that may contribute to more women misperceiving their symptoms is that, compared to men, more women had visited their primary doctor for symptoms related to an acute MI before being hospitalized (2). During these visits, women were more often told that their symptoms were not heart-related and instead were likely gastrointestinal or stress-related (2). If a patient explains to EMS that their doctor said their symptoms are gastrointestinal or stress-related, the prehospital provider may believe the patient and may not perform investigative interventions. Prehospital providers should be made aware of how patients may perceive their symptoms. This knowledge may enable EMS providers to overcome bias brought on by the patient’s symptoms and history.

Conscious or unconscious gender biases can affect a prehospital provider’s clinical decisions. Previous studies have found discrepancies in the performance of EMS interventions between male and female patients. For instance, a national study by Lewis et al. found that compared to men younger than 65 years old, women under 65 were significantly less likely to receive aspirin or nitroglycerin for chest pain (18). This study did not control for reasons not to administer these medications, such as patient administration before EMS arrival or an allergy (18). However, another possible reason not to administer medication for chest pain is a low suspicion of a cardiac event, which may be due to unconscious provider bias or inadequate training. Another observed discrepancy between women and men is that, for patients with chest pain, women under 65 were less likely to be transported using lights and sirens compared to men of similar age (18). Occasionally, a patient will request not to be transported with lights and sirens, but otherwise, lights and sirens are used to transport high-priority patients. These discrepancies in treatment and transport methods likely involve gender bias or disbelief of symptoms. Bringing awareness to discrepancies surrounding prehospital interventions may affect EMS providers’ patient care decisions, which ideally should be equally performed for women and men.

Failure to investigate the patient’s signs and symptoms could also lead to a delay in treatment. When paramedics suspect a cardiac emergency, they should perform an electrocardiogram (EKG). EKGs can show ST-elevated myocardial infarctions (STEMIs) or non-ST-elevated myocardial infarctions (NSTEMIs). The EKG reading can help paramedics assign a patient’s priority level, which EMS uses in their report while en route to their destination hospital. EMS providers assign patients experiencing a life-threatening medical emergency with a high priority. Once an EMS provider calls the destination hospital and gives a verbal report, the ED will prepare a room and the appropriate resources for high-priority patients before they arrive at the ED. Assigning a patient with a high priority likely decreases their time to imaging and/or treatment compared to patients assigned with lower priority. One concern that may arise is that basic life support crews (BLS), who are unable to conduct EKGs, may unnecessarily make an acute MI notification to the destination hospital based purely on signs and symptoms. The hospital may use unnecessary resources, and the patient may undergo avoidable stress from the experience. While it is best to send an advanced life support (ALS) crew to someone experiencing an acute MI, BLS crews may be sent if all other ALS crews in the area are busy. Nevertheless, both BLS and ALS providers should be aware of unfamiliar signs and symptoms of life-threatening disease states so that they may perform appropriate investigative interventions and decrease the possibility of delaying hospital evaluation.

During an acute MI, time to treatment can impact a patient’s outcome. Lichtman et al. found that women aged 18-65 had a longer time from symptom onset to hospital presentation (2). Deluca et al. found that each 30-minute delay from the onset of acute MI symptoms to a primary angioplasty was associated with both an increased relative risk of mortality in one year and an increased risk of having a pre-discharge ejection fraction of less than 30%.19 A normal ejection fraction is 52% to 72% for men and 54% to 74% for women (20). Per the American College of Cardiology, an ejection fraction below 30% is considered severely dysfunctional (20). A low ejection fraction indicates that the left ventricle is unable to pump blood to the rest of the body effectively, which can lead to further heart complications. Though EMS does not spend hours with a patient, they can potentially shorten the time to treatment if they recommend that the patient go to the hospital. Obtaining an EKG may especially benefit patients experiencing an MI who present with unfamiliar signs and symptoms and attribute their symptoms to other causes. 

In addition to provider recognition of a patient’s disease severity, social factors can also influence a patient’s time to treatment. Patient characteristics associated with prehospital delay include non-white race, low socioeconomic status, diabetes, and hypertension (21). A non-white race or low SES may be associated with a prehospital delay due to a lack of medical knowledge and mistrust in the healthcare system. Additionally, women are more likely to have a longer delay in seeking care (21). The reasons for the prolonged time to seek medical attention also tend to differ between women and men. Men who reported a mismatch between their expected and actual symptoms of an acute MI, had a low education level, did not call 911 or did not ride in an ambulance tended to have a prolonged time to seek treatment21. Meanwhile, women with a longer delay were older, single, alone during symptom onset, or did not want to trouble anyone (21). Interestingly, men with a history of an acute MI had a shorter delay in seeking treatment, while women with this history had a longer delay in seeking treatment (21). Though prehospital providers cannot impact these social factors, they potentially expedite getting the patient to definitive imaging and care. Suppose prehospital providers are aware of less common acute MI presentations. In that case, they may be able to advocate for a patient with a less common presentation and eliminate delays in the ED.

Conclusion –Curriculum Development and Future Directions

Since women are more likely to present with lesser-known signs and symptoms of acute MI and are less likely to receive the highest quality of care, there is a need to implement the above findings in prehospital education. This may be through classroom curriculum for future EMS providers and continuing education classes for current providers. EMS providers who understand that acute MI can present with lesser-known signs/symptoms may be more likely to obtain a detailed history and perform interventions, which subsequently may paint a clearer picture for emergency department providers. 

The goal of educating the prehospital provider about sex differences in acute MI is to increase awareness of how life-threatening conditions can manifest beyond the well-known textbook definitions and encourage providers to deliver higher-quality patient care. The curriculum should discuss trends in men’s and women’s clinical presentations and emphasize how women more often experience unfamiliar symptoms. To supplement this information, discussing underlying differences in male and female physiology could provide greater context for different trends in presentation. The physiology discussion should be tailored to either a more basic level for EMTs or a more detailed level for paramedics. The curriculum should also emphasize the current discrepancies in performing interventions and providing high-quality care between patient sexes. Due to these current discrepancies, the course should review EMS protocols for suspected acute MI. Additionally, the curriculum should reframe the language used to describe lesser-known signs and symptoms. Labeling the symptoms more commonly experienced by women as ‘atypical’ or ‘non-traditional’ diminishes their clinical importance. Emerging literature and curriculum should consider placing these signs and symptoms into a category that frames them as likely and valid disease indicators, such as ‘additional signs and symptoms.’

In addition to cisgender women and men, future research is needed to assess whether acute MI presents differently in non-cisgender individuals. About 1.6% of the U.S. population identifies as transgender or nonbinary, with a greater proportion under 30 years old (22). Transgender people may undergo hormone treatment or gender reconstruction surgery, and these procedures may affect the risk of developing acute MI. For instance, one study found that transgender men undergoing gender-affirming hormone therapy had an increased risk of MI compared to cisgender women and men (23). Transgender women undergoing gender-affirming hormone therapy may have an increased risk of an MI (23). Due to these increased risks, it may be helpful to compare the acute MI presentation of transgender populations to cisgender women and men. Another challenge that future research may face is collecting patient information. Transgender and non-binary patients do not always disclose their true gender identity to providers. For instance, the U.S. Transgender Survey reports that almost one-third of transgender people did not inform their healthcare providers that they were transgender (24). There are many reasons that people may not wish to disclose their gender, including prior discrimination, fear of mistreatment, and avoiding invasive questions (24). The transgender and non-binary individuals have especially experienced negative interactions with healthcare providers. The responsibility to make a patient feel comfortable falls on the medical provider, and, therefore, providers should have adequate training on how to care for transgender and nonbinary populations. The medical community must continue to acknowledge its biases and adjust providers’ approach to patient care so that patients of all genders may one day receive similar quality of care.

References

1. Geller SE, Koch AR, Roesch P, Filut A, Hallgren E, Carnes M. The More Things Change, the More They Stay the Same: A Study to Evaluate Compliance With Inclusion and Assessment of Women and Minorities in Randomized Controlled Trials. Acad Med. Apr 2018;93(4):630-635. doi:10.1097/ACM.0000000000002027

2. Lichtman JH, Leifheit EC, Safdar B, et al. Sex Differences in the Presentation and Perception of Symptoms Among Young Patients With Myocardial Infarction: Evidence from the VIRGO Study (Variation in Recovery: Role of Gender on Outcomes of Young AMI Patients). Circulation. Feb 20 2018;137(8):781-790. doi:10.1161/CIRCULATIONAHA.117.031650

3. Cardeillac M, Lefebvre F, Baicry F, et al. Symptoms of Infarction in Women: Is There a Real Difference Compared to Men? A Systematic Review of the Literature with Meta-Analysis. J Clin Med. Feb 27 2022;11(5)doi:10.3390/jcm11051319

4. Ali M, van Os HJA, van der Weerd N, et al. Sex Differences in Presentation of Stroke: A Systematic Review and Meta-Analysis. Stroke. Feb 2022;53(2):345-354. doi:10.1161/STROKEAHA.120.034040

5. Washington DL, Bird CE. Sex differences in disease presentation in the emergency department. Ann Emerg Med. Nov 2002;40(5):461-3. doi:10.1067/mem.2002.128859

6. Shajahan S, Sun L, Harris K, et al. Sex differences in the symptom presentation of stroke: A systematic review and meta-analysis. Int J Stroke. Feb 2023;18(2):144-153. doi:10.1177/17474930221090133

7. Jerath NU, Reddy C, Freeman WD, Jerath AU, Brown RD. Gender differences in presenting signs and symptoms of acute ischemic stroke: a population-based study. Gend Med. Oct 2011;8(5):312-9. doi:10.1016/j.genm.2011.08.001

8. Forster A, Gass A, Kern R, et al. Gender differences in acute ischemic stroke: etiology, stroke patterns and response to thrombolysis. Stroke. Jul 2009;40(7):2428-32. doi:10.1161/STROKEAHA.109.548750

9. The National Emergency Medical Services Education Standards (U.S. Department of Transportation, National Highway Traffic Safety Administration) (2021).

10. Dylla L, Rice JD, Poisson SN, et al. Analysis of Stroke Care Among 2019-2020 National Emergency Medical Services Information System Encounters. J Stroke Cerebrovasc Dis. Mar 2022;31(3):106278. doi:10.1016/j.jstrokecerebrovasdis.2021.106278

11. Haider A, Bengs S, Luu J, et al. Sex and gender in cardiovascular medicine: presentation and outcomes of acute coronary syndrome. Eur Heart J. Apr 01 2020;41(13):1328-1336. doi:10.1093/eurheartj/ehz898

12. Fahed AC, Jang IK. Plaque erosion and acute coronary syndromes: phenotype, molecular characteristics and future directions. Nat Rev Cardiol. Oct 2021;18(10):724-734. doi:10.1038/s41569-021-00542-3

13. Seegers LM, Araki M, Nakajima A, et al. Sex Differences in Culprit Plaque Characteristics Among Different Age Groups in Patients With Acute Coronary Syndromes. Circ Cardiovasc Interv. Jun 2022;15(6):e011612. doi:10.1161/CIRCINTERVENTIONS.121.011612

14. Iorga A, Cunningham CM, Moazeni S, Ruffenach G, Umar S, Eghbali M. The protective role of estrogen and estrogen receptors in cardiovascular disease and the controversial use of estrogen therapy. Biol Sex Differ. Oct 24 2017;8(1):33. doi:10.1186/s13293-017-0152-8

15. Kaur H, Werstuck GH. The Effect of Testosterone on Cardiovascular Disease and Cardiovascular Risk Factors in Men: A Review of Clinical and Preclinical Data. CJC Open. Oct 2021;3(10):1238-1248. doi:10.1016/j.cjco.2021.05.007

16. Lopez-Cuenca A, Gomez-Molina M, Flores-Blanco PJ, et al. Comparison between type-2 and type-1 myocardial infarction: clinical features, treatment strategies and outcomes. J Geriatr Cardiol. Jan 2016;13(1):15-22. doi:10.11909/j.issn.1671-5411.2016.01.014

17. Gupta S, Vaidya SR, Arora S, Bahekar A, Devarapally SR. Type 2 versus type 1 myocardial infarction: a comparison of clinical characteristics and outcomes with a meta-analysis of observational studies. Cardiovasc Diagn Ther. Aug 2017;7(4):348-358. doi:10.21037/cdt.2017.03.21

18. Lewis JF, Zeger SL, Li X, et al. Gender Differences in the Quality of EMS Care Nationwide for Chest Pain and Out-of-Hospital Cardiac Arrest. Womens Health Issues. Mar-Apr 2019;29(2):116-124. doi:10.1016/j.whi.2018.10.007

19. De Luca G, Suryapranata H, Ottervanger JP, Antman EM. Time delay to treatment and mortality in primary angioplasty for acute myocardial infarction: every minute of delay counts. Circulation. Mar 16 2004;109(10):1223-5. doi:10.1161/01.CIR.0000121424.76486.20

20. Kosaraju A, Goyal A, Grigorova Y, Makaryus AN. Left Ventricular Ejection Fraction. StatPearls. StatPearls Publishing Copyright © 2025, StatPearls Publishing LLC.; 2025.

21. Nguyen HL, Saczynski JS, Gore JM, Goldberg RJ. Age and sex differences in duration of prehospital delay in patients with acute myocardial infarction: a systematic review. Circ Cardiovasc Qual Outcomes. Jan 2010;3(1):82-92. doi:10.1161/CIRCOUTCOMES.109.884361

22. Brown A. <h1 aria-level="1" class="wp-block-post-title has-h-one-font-size" data-post-type="short-read" style="box-sizing: border-box; word-break: break-word; font-family: var(--wp--preset--font-family--serif); font-size: 35px; line-height: 41px; color: rgb(42, 42, 42); background-color: rgb(255, 255, 255);">About 5% of young adults in the U.S. say their gender is different from their sex assigned at birth. Pew Research Center. Accessed 7/22, 2024. 

23. Aranda G, Halperin I, Gomez-Gil E, et al. Cardiovascular Risk Associated With Gender Affirming Hormone Therapy in Transgender Population. Front Endocrinol (Lausanne). 2021;12:718200. doi:10.3389/fendo.2021.718200

24. James S.E. HJL, Rankin S., Keisling M., Mottet L.A., Anafi M. The Report of the 2015 U.S. Transgender Survey. 2016:92-129. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://transequality.org/sites/default/files/docs/usts/USTS-Full-Report-Dec17.pdf

Crystalline Conundrum: Understanding the Absence of Tumorigenesis in the Human Lens

Brianna C. Landis (1).

Affiliation
1. Rocky Vista University, College of Osteopathic Medicine, Ivins, Utah, USA.

Contact
255 East Center St. Ivins, UT 84738. Email: brianna.landis@rvu.edu.

Conflict of interest statement
The authors declare no potential conflicts of interest.

Abstract

Despite the mitotically active nature of the ocular lens and near-constant exposure to ultraviolet radiation, there have been no reported cases of primary tumors in the human lens. In contrast, such tumors have been induced and reported in the lens of non-human vertebrates, particularly in cats. This report discusses various theories, including the avascular nature of the lens, the presence of barrier properties within the ocular environment, and the lens capsule composition as a potential chemo-mechanical barrier against tumorigenesis. Despite the significant implications for cancer prevention and treatment, there has been limited research into this phenomenon. Identifying protective mechanisms could contribute to a better understanding of human cancer genetics and potentially lead to preventative treatments.

Significance statement

There are no reported cases of primary tumors within the human lens, a phenomenon that is poorly understood and scarcely investigated. This perspective highlights theories for potential tumor-resistant properties and urges researchers to continue investigating. Further research could have significant implications for understanding cancer biology, tumorigenesis, and future preventative treatments.

Over the past century, the absence of primary tumors of the human lens has been remarked upon in the literature but never rigorously investigated.1–4 If true, this observation is noteworthy and certainly remarkable because all dividing cells, even in invertebrates, can develop genetic mutations and form tumors.5 Immediately posterior to the anterior lens capsule lies a single layer of epithelial cells. Within this layer is the germinative zone. As cells divide, they migrate laterally and then displace centrally, losing organelles as they do so, so that they can laminate and incorporate into the clear crystalline lens that refracts light onto our retina (Figure 1).6 By design, this mitosis occurs throughout life and into old age, thus it is curious that such a mitotically active tissue, with near constant exposure to ultraviolet radiation, is seemingly resistant to tumorigenesis.

Figure 1. The mitotic cycle of epithelial cells within the normal human crystalline lens (6). Image created using BioRender.com.

Limited research has explored this phenomenon, the results of which are equally perplexing. Malignant tumors of the lens can be induced with exposure to carcinogens (7) and oncogenic viruses (8) and can be engineered genetic defects in transgenic animals (9–11). Strikingly, malignant tumors of the lens can occur spontaneously in other nonhuman vertebrate species (cats, rabbits, dogs, and birds) (12–20). 

However, no case of malignant or benign tumor of the human lens has been reported or described in the literature. A review of veterinary databases revealed that in non-human species, malignant tumors of the lens occur most commonly in cats, constituting 4.5% of intraocular and adnexal neoplasms in that species (20). It is established that rupture of the lens capsule is a major risk factor for the development of the tumor, coining the nomenclature of feline ocular post-traumatic sarcoma (a benign neoplasm of lens epithelial origin) (14). Retrospective review of previously unreported cases of primary lens tumors in cats from the University of Wisconsin School of Veterinary Medicine's Comparative Ocular Laboratory collection reveals that all cases showed evidence of lens capsule rupture and most had some degree of uveitis, similar to tumors observed in other vertebrate species (birds, rabbit, and dog) (20). 

With awareness of the strong correlation of capsular trauma to tumorigenesis, it is even more shocking that such tumors have not been described in humans. Cataract removal is the most commonly performed surgical procedure in humans (21), and by design, the lens epithelium is traumatized and retained during modern extracapsular cataract extraction. Despite frequent surgical injury to the lens epithelium, no benign or malignant lens tumors have been documented (20). Could genetic protective mechanisms exist? And if so, wouldn’t identifying this mechanism hold significant implications to better understanding the genetics of human cancers and ultimately providing preventative treatment?

Several theories have arisen to attempt to explain this phenomenon. The lens is a naturally avascular tissue, sensibly designed to minimize the scattering of light as it passes through to the retina. As such, the lens acquires nutrients from aqueous and vitreous components passing through the semipermeable membrane of the lens capsule (22). It has long been proposed and accepted that adequate vascular supply is essential for tumorigenesis and progression (23). Solid tumors, irrespective of their source, typically begin as a small cluster of cells relying on nutrients diffusing from nearby tissues (23). As the tumor grows, it eventually reaches a size where simple diffusion is inadequate to support further growth and angiogenesis is needed to facilitate further growth. Hence, it is sensible to assume an avascular tissue, such as the lens, could harbor potential tumors to a minimal size dependent on available resources. Interestingly, however, even proangiogenic colonies of neoplastic cells have not been described (24). 

Additionally, another avascular ocular tissue, the cornea, can still be invaded by advancing tumors maintaining their angiogenic factors (25). As such, avascularity alone does not entirely explain the lack of primary tumor formation in the lens. However, unlike the lens, the cornea lacks a capsular barrier. Although, it is hypothesized that the Bowman’s layer of the cornea serves as a form of corneal barrier, as tumors in the stroma layer beneath are largely undiscovered despite a notable prevalence of chromosomal abnormalities (26). Hence, it is conceivable that either the ocular environment or the existence of barrier properties could contribute to the absence of tumor development in the lens.

Other theories emphasize this possibility by suggesting the lens capsule is a chemo-mechanical barrier. Among other molecules, the lens capsule is largely composed of collagen types I-IV (27, 28). Fragments of collagens make up endostatins, which are known to act as inhibitors of angiogenesis.29 It is postulated that these endostatin molecules exist near or within the lens capsule to serve a protective mechanism against angiogenesis, both for the purpose of preserving lens transparency but also inhibiting the angiogenesis of tumors (29). Notably, there is evidence indicating that fragments of type IV collagen, the primary constituent of the lens capsule, may impede tumor cell growth (30) and hinder the activation of matrix metalloproteinases (31) in tumor cells believed to contribute to invasiveness. This evidence lends support to the idea that a growth inhibitor associated with the lens capsule could prevent neoplastic transformation in subcapsular epithelial cells.

Interestingly, even highly invasive melanomas and retinoblastomas, sometimes occupying the entirety of the posterior chamber, demonstrate well-defined borders at the lens capsule interface (24). These types of tumors are widely recognized for their infrequent invasion or direct contact with the lens capsule. Instead, the tumor-lens interface becomes filled with debris and fluid (24). Is it possible that these tumors are repelled by an unknown chemo-mechanical property of the lens capsule?

Undeniably, these observations are quite remarkable and impress exciting potential for research to advance cancer prevention and treatment. So then why is there such limited research into this phenomenon? According to NIH.gov, in 2020, the National Institute of Health (NIH), a major research funding agency in the United States, estimated a cost of nearly 6.5 billion dollars to support cancer research efforts. However, a search into the NIH RePORTER database revealed that the NIH has never funded any projects attempting to investigate the seemingly tumor-resistant properties of the human lens capsule epithelium (32). Such research seems to be a prime candidate for identifying a cancer-inhibiting gene or genes in humans. With current genetic techniques, it should be possible to identify the genes involved in lens tumor formation in other species and to use these as candidate genes in identifying the genes responsible for preventing cancer in the human lens. A genetic protective mechanism is hypothesized to exist, and identifying this mechanism may be of significant value in enhancing the understanding of human cancer genetics.

Given the striking absence of primary tumors in the human lens, several experimental approaches could be employed to elucidate the underlying protective mechanisms, such as comparative genomic transcriptomic analysis, CRISPR-based functional genomics, lens capsule extracellular matrix components, in vivo animal models, epigenetic and regulatory network studies, and organoid and 3D cell culture models (Table 1).

Table 1. Potential experimental methodologies and genetic approaches for investigating protective mechanisms of the human lens and their associated research benefits.

By employing these methodologies, researchers could move beyond theoretical explanations and begin identifying actionable molecular targets for preventing tumorigenesis in other tissues. Understanding how the lens naturally resists tumor formation could lead to the development of novel cancer therapies, including biomimetic extracellular matrices, anti-angiogenic compounds, and gene therapies designed to enhance tumor suppression in high-risk tissues.

The apparent resistance of the human lens epithelium to tumorigenesis presents an untapped avenue for oncological research, with potential implications for cancer prevention and treatment. The absence of primary tumors in a tissue that remains mitotically active throughout life, despite continuous exposure to ultraviolet radiation and surgical trauma, suggests the existence of unique protective mechanisms. If genetic or biochemical factors within the lens epithelium or capsule contribute to this resistance, identifying these mechanisms could inform strategies for suppressing tumorigenesis in other tissues. For example, if the lens capsule's extracellular matrix components, such as type IV collagen and endostatins, play a role in inhibiting angiogenesis and tumor invasion, similar mechanisms might be leveraged therapeutically in cancers reliant on angiogenic signaling. Moreover, uncovering genetic factors that prevent neoplastic transformation in the lens could contribute to the identification of novel tumor suppressor genes, expanding our understanding of intrinsic cancer resistance in humans. This phenomenon underscores the necessity of further research, as elucidating these protective mechanisms could inspire innovative approaches to cancer prevention and treatment, shifting the focus from reactive therapies to proactive, biologically informed interventions.

Acknowledgments

The author would like to acknowledge Anthony Pappas, Ph.D. for introducing this fascinating lapse in knowledge, as well as student doctors Parker Webber and Bosten Loveless for their contributions to the poster presentation Paradoxical lack of investigation into the natural tumor-resistant properties of the human lens capsular epithelium. Their efforts were paramount in sparking interest in this subject and highlighting this notable gap in the literature.

References

1. Sachs E, Larsen RL. Cancer and the Lens. Am J Ophthalmol. 1948;31(5):561-564. doi:10.1016/0002-9394(48)90558-3

2. Ullrich K, Casson RJ. Does anybody care that the crystalline lens never gets cancer? Clin Experiment Ophthalmol. 2013;41(8):812-812. doi:10.1111/ceo.12114

3. Mann I. Induction of an Experimental Tumour of the Lens. Br J Ophthalmol. 1947;31(11):676-685. doi:10.1136/bjo.31.11.676

4. M. Seigel G. The enigma of lenticular oncology. Digit J Ophthalmol. 2001;7(4). Accessed December 13, 2023. https://legacy.djo.harvard.edu/site.php%3Furl=%252Fphysicians%252Foa%252F360.html

5. Domazet-Lošo T, Klimovich A, Anokhin B, et al. Naturally occurring tumours in the basal metazoan Hydra. Nat Commun. 2014;5(1):4222. doi:10.1038/ncomms5222

6. Lens and Cataract, Chapter 2: Anatomy. In: 2020–2021 BCSC Basic and Clinical Science Course. American Academy of Ophthalmology. Accessed December 13, 2023. https://www.aao.org/education/bcscsnippetdetail.aspx?id=298fbd36-d41e-4714-80d9-f55e41ec4624

7. Von Sallmann L, E. Halver J, Collins E, Grimes P. Thioacetamide-induced Cataract with Invasive Proliferation of the Lens Epithelium in Rainbow Trout. Cancer Res. 1966;26:1819-1825.

8. Albert DM, Rabson AS, Grimes PA, Von Sallmann L. Neoplastic Transformation in vitro of Hamster Lens Epithelium by Simian Virus 40. Science. 1969;164(3883):1077-1078. doi:10.1126/science.164.3883.1077

9. Chen Q, Hung FC, Fromm L, Overbeek PA. Induction of cell cycle entry and cell death in postmitotic lens fiber cells by overexpression of E2F1 or E2F2. Invest Ophthalmol Vis Sci. 2000;41(13):4223-4231.

10. Zheng H chuan, Nakamura T, Zheng Y, et al. SV40 T antigen disrupted the cell metabolism and the balance between proliferation and apoptosis in lens tumors of transgenic mice. J Cancer Res Clin Oncol. 2009;135(11):1521-1532. doi:10.1007/s00432-009-0599-z

11. Mahon KA, Chepelinsky AB, Khillan JS, Overbeek PA, Piatigorsky J, Westphal H. Oncogenesis of the Lens in Transgenic Mice. Science. 1987;235(4796):1622-1628. doi:10.1126/science.3029873

12. Woog J, Albert DM, Gonder JR, Carpenter JJ. Osteosarcoma in a Phthisical Feline Eye. Vet Pathol. 1983;20(2):209-214. doi:10.1177/030098588302000208

13. Dubielzig RR. Ocular sarcoma following trauma in three cats. J Am Vet Med Assoc. 1984;184(5):578-581.

14. Dubielzig RR, Everitt J, Shadduck JA, Albert DM. Clinical and Morphologic Features of Post-traumatic Ocular Sarcomas in Cats. Vet Pathol. 1990;27(1):62-65. doi:10.1177/030098589002700111

15. Wong CJ, Peiffer RL, Oglesbee S, Osborne C. Feline ocular epithelial response to growth factors in vitro. Am J Vet Res. 1996;57(12):1748-1752.

16. Zeiss CJ, Johnson EM, Dubielzig RR. Feline intraocular tumors may arise from transformation of lens epithelium. Vet Pathol. 2003;40(4):355-362. doi:10.1354/vp.40-4-355

17. Carter RT, Giudice C, Dubielzig RR, Colitz CMH. Telomerase Activity with Concurrent Loss of Cell Cycle Regulation in Feline Post-traumatic Ocular Sarcomas. J Comp Pathol. 2005;133(4):235-245. doi:10.1016/j.jcpa.2005.04.009

18. McPherson L, Newman SJ, McLean N, et al. Intraocular Sarcomas in Two Rabbits. J Vet Diagn Invest. 2009;21(4):547-551. doi:10.1177/104063870902100422

19. Dickinson R, Bauer B, Gardhouse S, Grahn B. Intraocular sarcoma associated with a rupture lens in a rabbit ( Oryctolagus cuniculus ). Vet Ophthalmol. 2013;16(s1):168-172. doi:10.1111/vop.12049

20. M. Albert D, O. Phelps P, R. Surapaneni K, et al. The Significance of the Discordant Occurrence of Lens Tumors in Humans versus Other Species. Ophthalmology. 2015;122(9).

21. March 2015 Report to the Congress: Medicare Payment Policy – MedPAC. Accessed December 13, 2023. https://www.medpac.gov/document/http-www-medpac-gov-docs-default-source-reports-mar2015_entirereport_revised-pdf/

22. H. D. The Lens. In: Physiology of the Eye. 5th ed. New York Pergamon Press; :145-149.

23. Folkman J, Merler E, Abernathy C, Williams G. Isolation of a tumor factor responsible for angiogenesis. J Exp Med. 1971;133(2):275-288.

24. Chaturvedi S, Mehrotra AN, Mittal S, Bahadur H. The Conundrum of Lenticular Oncology. A Review. Indian J Ophthalmol. 2003;51(4):297.

25. Langer R, Brem H, Falterman K, Klein M, Folkman J. Isolations of a Cartilage Factor That Inhibits Tumor Neovascularization. Science. 1976;193(4247):70-72. doi:10.1126/science.935859

26. Pettenati MJ, Sweatt AJ, Lantz P, et al. The human cornea has a high incidence  of acquired chromosome abnormalities. Hum Genet. 1997;101(1):26-29. doi:10.1007/s004390050580

27. Marshall GE, Konstas AGP, Bechrakis NE, Lee WR. An immunoelectron microscope study of the aged human lens capsule. Exp Eye Res. 1992;54(3):393-401. doi:10.1016/0014-4835(92)90051-S

28. Schmut O. The organization of tissues of the eye by different collagen types. Albrecht Von Graefes Arch Für Klin Exp Ophthalmol. 1978;207(3):189-199. doi:10.1007/BF00411053

29. Sasaki T, Larsson H, Tisi D, Claesson-Welsh L, Hohenester E, Timpl R. Endostatins derived from collagens XV and XVIII differ in structural and binding properties, tissue distribution and anti-angiogenic activity11Edited by A. Fersht. J Mol Biol. 2000;301(5):1179-1190. doi:10.1006/jmbi.2000.3996

30. Kefalides NA, Monboisse JC, Bellon G, Ohno N, Ziaie Z, Shahan TA. Suppression of tumor cell growth by type IV collagen and a peptide from the NC1 domain of the alpha 3(IV) chain. Medicina (Mex). 1999;59(5 Pt 2):553.

31. Pasco S, Han J, Gillery P, et al. A specific sequence of the noncollagenous domain of the alpha3(IV) chain of type IV collagen inhibits expression and activation of matrix metalloproteinases by tumor cells. Cancer Res. 2000;60(2):467-473.

32.Webber P, Landis B, Loveless B, C Pappas A. Paradoxical lack of investigation into the natural tumor-resistant properties of the human lens capsular epithelium. Invest Ophthalmol Vis Sci. 2023;64(8):4140.

A Guide to AI Challenges and Barriers

Rosemarie Burynski (1); Bernice L. Hausman (2).

Affiliations
1. Penn State College of Medicine, Hershey, Pennsylvania.
2. Department of Humanities, Penn State College of Medicine, Hershey, Pennsylvania.

Contact
Corresponding author: R.B.: rburynski@pennstatehealth.psu.edu

Conflict of interest statement
We have no disclosures or conflicts of interest.

Abstract

Artificial intelligence (AI) tools are developing quickly and prominently within the U.S healthcare system. It therefore seems essential to understand their weaknesses in order to practice medicine that is fully informed. The goal of this paper is to overview several key concerns surrounding healthcare AI, as well as some anticipated barriers to its implementation. AI’s generalizability is currently limited due to a widely fragmented Electronic Health Record (EHR) and inaccessibility to training data. AI tools are therefore at risk of acting on incomprehensive knowledge and generating inaccurate outputs. They are also extremely susceptible to several different forms of bias. Such bias can result in preferences towards diagnosing some diseases over others, and recommending interventions that are only beneficial to certain populations. Privacy and transparency are also of great concern, especially when dealing with private medical data. While “black box” algorithms are criticized for their lack of transparency, innovators are working towards explainable AI (XAI) tools that can “show their work.” Developing guidelines make it difficult to predict how liability for AI malpractice may be distributed across parties, but has interesting implications for how physicians will change their practice in response. Finally, the current U.S. payment structure does not easily accommodate healthcare AI tools. This challenge raises questions surrounding healthcare AI’s reimbursement mechanism as it becomes more widely utilized. While this paper does not provide solutions for the outlined concerns, it emphasizes the importance of understanding and anticipating the shortcomings of new healthcare technologies.

Introduction 

As artificial intelligence (AI) becomes increasingly prevalent, so do concerns regarding its ability to accurately and equitably supplement the medical field. Therefore, it is the duty of providers to be aware of both the benefits and harms that healthcare AI may pose towards their patients. This guide overviews some major talking points surrounding healthcare AI’s anticipated challenges and implementation barriers.

 Generalizability

            A major concern within the field of AI research is the generalizability of results. There are a few reasons for this. One is that there is no universal Electronic Health Record (EHR) within the U.S. Most current healthcare AI is supervised, meaning that it requires training on large data sets that are labeled by humans. This training is difficult to accomplish when health data is scattered throughout different systems. Many programs must settle for smaller amounts of training data. This limitation raises questions regarding the applicability of those systems to larger or different populations of patients. This concern is also prevalent on a smaller scale - AI with great performance using data from one U.S. hospital may fail in another U.S. hospital due to its lack of generalizability (1). One exception to the fragmented American EHR is the VA, the country’s biggest integrated healthcare system (2). The vast amount of data stored within the VA EHR makes for an ideal AI training ground. However, the data demographics still leave plenty of room for debate - How generalizable is VA data to the rest of the country? 

AI training seems to risk spectrum bias, which refers to a test that is performed and evaluated within a population that is different from the intended population (3). This incomprehensive training leads to limited, incomprehensive knowledge. If medical students are only taught to recognize and treat diseases that are prevalent in their school’s state, they will misdiagnose and mistreat the vast array of diseases that are less geographically common. If they are taught only to identify infections on lighter skin tones, they are more likely to miss presentations on darker skin tones. In fact, they may fail to identify the presence of disease altogether because they have had no exposure to it in training. The same problem applies to AI.

 Other Biases

         In general, a program that is trained on biased data will inherit and reinforce inequalities in its own algorithms. Echoing the concerns of generalizability is diagnosis bias. COVID-19 diagnostic tools that are trained in the U.S. may not have much exposure to lung-related diseases such as tuberculosis and types of pneumonia associated with HIV/AIDS that are more prevalent in other countries. The algorithm then runs the risk of misdiagnosing these diseases as COVID-19 due to their similarities and lack of knowledge of their differences (4).

            Bias may also be introduced in disease modeling scenarios. A big mitigation initiative during the peak of the COVID-19 pandemic was disease “mapping” to track spread. These modeling techniques often require specified data inputs that are more challenging to obtain from underrepresented populations. Additionally, the models were used in recommending interventions (such as quarantining and social distancing) that were a lot less attainable in crowded and/or poor sanitary environments. Similarly, treatment selections made through AI tools were less likely to account for social determinants of health that are underrepresented and less accurately documented in the EHR (4).

            Infodemic bias is especially prevalent in the age of social media and mass information. AI has increasingly been used to help fact-check and combat the spread of misinformation. However, this integration is mostly done within “easy to mine” data sources such as Twitter and Facebook. While it may be having a positive impact on these platforms, other information sources such as radio and TV are less likely to be fact-checked by AI. Simultaneously, these alternate channels may act as primary information sources for certain countries and populations (4).

 Privacy, Transparency and Mistrust

            By design, medical AI is going to guide and influence clinical decision-making. As always, it is important for a physician to be able to explain how and why a recommendation is made. This process is made more complex when considering the “black box” tendencies of certain AI technologies. The term “black box” is used to refer to the lack of transparency regarding AI output. It is not always as clear how an AI algorithm generated a particular output given a specific input. This lack of clarity can lead to an overall mistrust of AI, and especially so within the healthcare field. How can a physician use the medical advice of an algorithm when its reasoning is not available? 

            These concerns have led to the push for explainable AI (XAI) that is able to reason through its output in an understandable way. There are a wide variety of XAI methods, many of which include visual representations (decision trees, graphs, etc.) of the decision-making process (5). These methods ensure that when a clinician inputs data, they receive not only the output they are looking for but also an explanation for that output. The clinician can use this information in their own professional evaluation of the AI’s performance before utilizing its advice in their decision-making process. Given the extreme importance of physician transparency, it will be unsurprising to see XAI continue to grow throughout the field.

Rules, Regulations, and Malpractice

AI has been leveraged by developers through the promise of its increased accuracy, and hence, its ability to reduce medical mistakes. However, the rules and regulations surrounding responsibility for AI mishaps are still developing. Historically, physicians are typically liable for their actions even when under the influence of third party information. For example, if a physician follows an insurer’s recommendation for a procedure plan and harm occurs, the physician is still responsible. Similarly, the physician is responsible for appealing coverage denials if they believe a service to be medically necessary (2). In the case of AI, then, it would seem that physicians should maintain liability for all decision-making. However, the lack of clear guidelines may still leave room for liability to be potentially shared by AI producers. This possibility combined with the hope of increased accuracy puts malpractice in an interesting economic position.

On the one hand, malpractice pressure increases the demand for AI. Physicians may shift towards “defensive medicine,” through which they’d push certain decision-making onto AI in hopes of avoiding some amount of liability. This increased demand for AI would result in raised AI prices.Competition increases and product differentiation decreases. Subsequently, AI producers would have to rely more on prices in order to compete and would likely reduce their prices in response. Therefore, malpractice seemingly has two opposite effects on price-setting and profit-making6. While the long-term economic effect is still up for questioning, it does certainly raise social concerns regarding how malpractice will affect physician reliance on AI.

What about insurance?

         The already complicated world of health insurance is made more complex when considering how AI can or should be billed for. In general, medical procedures and services are defined by Current Procedural Terminology (CPT) codes that are developed by the American Medical Association (AMA). These codes are divided into three categories (7):

1. Category I → Describes a procedure or service that must meet specific criteria. These codes are typically reimbursed by both Medicare and commercial payers.

2. Category II → Used for tracking and performance measurement purposes. These codes are not generally reimbursed by Medicare or commercial payers.

3. Category III → Codes for developing technology, services, and procedures. These codes are temporary and may be later placed in Category I if the criteria is met. While there are no fees assigned to these codes, reimbursement may be available on a case-by-case basis (8).

To be billed for, a specific AI technology must fall into a CPT category and be defined by a CPT code. AI does not fit well into this type of payment structure. New AI technologies are performing countless and various tasks. It would take a long and tedious amount of time to create a special CPT code for each (7). Simultaneously, one CPT code cannot overlap with another. This rule provides a unique challenge for AI, as many algorithms do work that has historically been performed by humans and is likely already defined by an existing CPT code. For example, an AI tool that detects pulmonary hypertension in medical imaging is performing work that is already covered by CPT code CPT71275: CT angiography, chest (noncoronary) w/ contrast material(s), including noncontrast images, if performed, and image postprocessing (7). One suggestion has been to bundle AI services with their complementary services (in this case, the new pulmonary hypertension AI tool and the already-existing imaging service would be bundled) (9). However, it is still unclear as to what billing trajectory AI will end up following.

Conclusion

The rapid rise of healthcare AI promises big change for the medical field. The associated challenges outlined in this paper are non-exhaustive and subject to change over time. Still, understanding them will guide learning and decision-making when implementing new AI tools into the healthcare system.

References

1. Cossio, M., & Gilardino, R. E. (2021). Would the Use of Artificial Intelligence in COVID-19 Patient Management Add Value to the Healthcare System? Frontiers in Medicine, 8. https://doi.org/10.3389/fmed.2021.619202

2. Agrawal, A., Gans, J., Goldfarb, A., & Tucker, C. (2024). The Economics of Artificial Intelligence. University of Chicago Press.

3. Gupta, A., Slater, J., Boyne, D. J., Mitsakakis, N., Béliveau, A., Drużdżel, M. J., Brenner, D. R., Hussain, S., & Arora, P. (2019). Probabilistic Graphical Modeling for Estimating Risk of Coronary Artery Disease: Applications of a Flexible Machine-Learning Method. Medical Decision Making, 39(8), 1032–1044. https://doi.org/10.1177/0272989x19879095

4. Luengo-Oroz, M., Bullock, J., Pham, K. H., Lam, C. S. N., & Luccioni, A. (2021). From Artificial Intelligence Bias to Inequality in the Time of COVID-19. IEEE Technology and Society Magazine, 40(1), 71–79. https://doi.org/10.1109/mts.2021.3056282

5. Sarp, S., Catak, F. O., Kuzlu, M., et al (2023). An XAI approach for COVID-19 detection using transfer learning with X-ray images. Heliyan, 9(4), e15137–e15137. https://doi.org/10.1016/j.heliyon.2023.e15137

6. Chopard, B., & Musy, O. (2023). Market for artificial intelligence in health care and compensation for medical errors. International Review of Law and Economics, 75, 106153. https://doi.org/10.1016/j.irle.2023.106153

7. Smetherman, D., Golding, L., Moy, L., & Rubin, E. (2022). The Economic Impact of AI on Breast Imaging. Journal of Breast Imaging, 4(3), 302–308. https://doi.org/10.1093/jbi/wbac012

8. Dotson, P. (2013). CPT® Codes: What Are They, Why Are They Necessary, and How Are They Developed? Advances in Wound Care, 2(10), 583–587. https://doi.org/10.1089/wound.2013.0483

9. Zink, A., Chernew, M. E., & Neprash, H. T. (2024). How Should Medicare Pay for Artificial Intelligence? JAMA Internal Medicine, 184(8). https://doi.org/10.1001/jamainternmed.2024.1648

Forgotten Fragments

Shalini Radhakrishnan [1]
[1] Department of Pathology, Kasturba Medical College of Manipal, Managlore, India
Correspondance: Shalini.Radhakrishnan@learner.manipal.edu

When words decayed and thoughts grew dim,

I lost my grip on life’s fragile limb

Memory, once a symphony, now a lost hymn,

And darkness seeped beneath my paper-thin skin.

The hours withered like a barren vine,

Leaving me floating on a sea of time,

My recollections vanished, line by line,

Like ink-tipped memories drained by a thirsty pen of mine.

I became a pile of forgotten keys,

Shrouded in a cloak of elusive memories,

A whirlwind of tangled threads and mysteries,

Binding my mind, unraveling life’s sweet melodies.

Names and faces now distant shores,

Whose norm I could perceive no more,

Fractured fragments of a life once bore,

Collapsed like a forgotten tapestry of yore.

The man I knew, a mirage of mist,

Drifting through the layers of a life amiss,

A puzzle box with an erased life list,

Each day a ghost, a specter I resist.

My family recoiled, strangers in my gaze,

No longer found in the labyrinths of my haze,

An empty mirror reflecting a vacant maze,

The essence of me, lost in a forgotten haze.

I wandered down corridors of vacant thought,

A haunted mansion of memories unsought,

Whispering echoes of who I once sought,

Lost in the labyrinth of my mind, distraught.

The world a riddle, my mind the key,

Yet I fumble and stumble, unable to see,

A ravenous hunger for clarity consumes me,

But the answers are locked in a forgotten decree.

I long for the embrace of yesterday’s light,

To reclaim my thoughts, my essence, my might,

But I’m trapped in a twilight so tragically bright,

Where shadows dance and memories take flight.

In this fractured world, I drift and sway,

A vessel adrift in a turbulent bay,

My thoughts like fragments, drifting away,

Lost in the ether, forever to stay.

But in spite of the pain, the loss, the grief,

In the darkest depths, I find a slight relief,

For even in fragments, there’s still belief,

That traces of me may someday find relief.