Letters Figure. Primary and Secondary Implantable Cardioverter Defibrillator (ICD) Use, Before and After the Mandate (2016-2020) |~Ä~| Primary preventio • •• • • • AS • • November September July 2015 2016 2017 May March January 2018 2019 2020 Secondary prevention •• • • November September 2015 2016 July 2017 May March January 2018 2019 2020 Each observation represents the number of de novo ICD implantations per lOOOOO Medicare beneficiaries eligible for a primary-prevention (A) or secondary-prevention (B) ICD per month. The vertical dashed line represents the month the shared decision-making mandate went into effect. The solid lines are the result of ordinal-least squares regression models using Newey-West standard errors to adjust for the autocorrelation and heteroskedasticity of time-series data. The shaded areas represent the 95% CIs. There is no statistical difference between the premandate trends of the primary- and secondary-prevention groups (estimated difference, 0.19 procedures per 100 000 per month; 95% CI, -1.68 to 2.06; P = .84). Limitations include the inability to ascertain key patient factors to determine eligibility for ICDs (eg, ejection fraction); thus, we may have overestimated the number of eligible patients. Furthermore, we could not explore whether this policy resulted in more frequent SDM for ICDs due to a lack of routine tracking of this outcome. Future policy aimed at increasing patient-centered care through SDM should identify the key outcomes that the policy aims to improve and incentivize clinical activities that enable tracking of these outcomes. Joshua B. Rager, MD, MA, MS Hechuan Hou, MS Tanner Caverly, MD, MPH Michael P. Thompson, PhD, MPH Author Affiliations: National Clinician Scholars Program, University of Michigan, Ann Arbor (Rager); Department of Cardiac Surgery, University of Michigan, Ann Arbor (Hou, Thompson); Division of General Medicine, Department of Internal Medicine, University of Michigan, Ann Arbor (Caverly). Accepted for Publication: December 10,2023. Published Online: February 19,2024. doi:10.1001/jamainternmed.2023.8532 Corresponding Author: Joshua B. Rager, MD, MA, MS, University of Michigan, 2800 Plymouth Rd, North Campus Research Complex, Bldg 14, Room G100-16, Ann Arbor, MI48109-2800Gbrager@med.umich.edu). Author Contributions: Dr Rager had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Conceptond design: Rager, Thompson. Acquisition, analysis, or interpretation of data: Ml authors. Drafting of the manuscript: Rager, Thompson. Critical review of the manuscript for important intellectual content: Ml authors. Statistical analysis: Rager, Hou. Obtained funding: Thompson. Administrative, technical, or material support: Rager, Thompson. Supervision: Caverly, Thompson. Conflict of Interest Disclosures: Dr Caverly reported having a license for a decision support tool for lung cancer screening (screenLC.com; licensed to Tanner Caverly and Angie Fagerlin; Apache 2.0 license; open source). Dr Thompson reported receivinggrants from the Agency for Healthcare Research and Quality and Blue Cross Blue Shield of Michigan outside the submitted work. No other disclosures were reported. Funding/Support: Support for this project was provided by the Department of Veterans Affairs' Office of Academic Affiliations Advanced Fellowships and by the University of Michigan National Clinician Scholars Program. Role of the Funder/Sponsor: Thefunders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. Data Sharing Statement: See Supplement 2. 1. Jensen TS, Chin J, AshbyL, Dolan D, CanosD, Hutter J. Implantable Cardioverter Defibrillators. CMS.gov. Published February 15,2018. Accessed December 30,2022. https://www.cms.gov/medicare-coverage-database/view/ ncacal-decision-memo.aspx?proposed=N&NCAId=288 2. Oshima Lee E, Emanuel EJ. Shared decision making to improve care and reduce costs. N£ng/JMed.2013;368(1):6-8.doi:10.1056/NEJMp1209500 3. El-Chami MF, Jacobsen CM, Griffiths Rl, et al. Device-related infection in de novo transvenous implantable cardioverter-defibrillator Medicare patients. Heart Rhythm. 2021;18(8):1301-1309. doi:10.1016/j.hrthm.2021.04.014 4. Veroff D, MarrA, WennbergDE. Enhanced support for shared decision making reduced costs of care for patients with preference-sensitive conditions. Health Aff (Millwood). 2013;32(2):285-293. doi:10.1377/hlthaff.2011.0941 5. Hurley VB, Rodriguez HP, Kearing S, Wang Y, Leung MD, Shortell SM. The impact of decision aids on adults considering hip or knee surgery. Health Aff (Millwood). 2020;39(1):100-107. doi:10.1377/hlthaff.2019.00100 6. Merchant FM, Dickert NW Jr, Howard DH. Mandatory shared decision making by the Centers for Medicares Medicaid Services for cardiovascular procedures and other tests. JAMA. 2018;320(7):641-642. doi:10.1001/jama. 2018.6617 Injuries From Legal Interventions Involving Conducted Energy Devices Police departments use conducted energy devices (CEDs), such as TASERs (TASER Self-Defense), as less lethal alternatives to firearms. With CEDs, compressed nitrogen charges propel metal barbs with wires that implant into the target. Electrical pulses up to 50 000 V are transmitted through barbs, causing incapacitation and loss of neuromuscular control.1 In 2019, International Statistical Classification of Diseases and Related Health 440 JAMA Internal Medicine April 2024 Volume 184, Number4 © 2024 American Medical Association. All rights reserved. jamainternalmedicine.com Letters Table 1. Characteristics of Law Enforcement-Related Conducted Energy Device Injuries by Body Region* Body region Characteristic Overall Abdomen Face Chest Head and neck Extremities General No. of patients with injuries (%)b 904 (70.8) 224(24.2) 73 (8.7) 256(27.4) 224 (25.6) 329 (36.9) 43 (5.0) Weighted %of patients by injury typec Abrasion 28.0 20.7 49.1 23.0 40.1 40.7 3.4 Concussion 1.2 NA NA NA 4.5 0.4 NA Contusion 12.0 7.6 20.2 10.0 22.1 16.1 4.0 Foreign body 10.1 14.0 1.7 14.7 4.1 8.9 NA Fracture 4.3 NA 27.7 2.8 6.5 4.6 NA Laceration 13.2 6.5 34.0 9.0 23.7 14.7 2.4 Pain 10.2 10.4 11.9 7.1 14.9 12.7 10.1 Punctures 31.4 54.3 12.5 48.0 13.2 30.7 5.3 TBI 0.3 NA 1.3 NA 1.4 NA NA Unspecified 17.4 11.3 8.8 14.0 23.9 11.1 93.5 Weighted % by maximum AIS severity 1: Minor 61.1 61.8 67.1 53.0 51.8 71.1 3.9 2: Moderate 25.4 20.3 22.9 30.2 37.6 20.2 45.3 3: Serious 9.0 6.1 8.6 12.8 6.6 6.0 50.7 4: Severe 2.9 10.8 NA 1.3 0.5 1.8 NA 5: Critical 1.6 1.0 1.4 2.8 3.5 0.9 NA Abbreviations: AIS, Abbreviated Injury Scale; NA, not applicable; TBI, traumatic body region. The denominator includes all injured patients, brain injury. c |nc|icates the number of injuries by type and conditional percentages among 'Authors' analysis of Nationwide Emergency Department Sample (NEDS). All patients who sustained an injury to the specified body region. Patients may percentages were weighted by NEDS discharge weights. have multiple injury types and body regions reported; thus, the number of b Percentages indicate the likelihood of sustainingan injury to the specified region-specific injuries may exceed overall totals. Problems, Tenth Revision (ICD-IO) codes were added to indicate CED use by law enforcement. We evaluated sociodemo-graphic and clinical characteristics of patients presenting with law enforcement-related CED injuries. Methods | We sampled US emergency department (ED) visits with ICD-IO code Y35.83X from the 2019-2020 Nationwide Emergency Department Sample (NEDS),2 which provided a 20% stratified sample of all EDs and weights to allow cal-Editor's Note page 373 culation of nationally repre. sentative estimates for all ED Related article page 363 visits. Data included patient sociodemographic character-Supplemental content istics, clinical features of encounters, and institutional characteristics. Unit of analysis was the individual ED visit. Vanderbilt University Medical Center Institutional Review Board deemed this cross-sectional study exempt from review. Organizations participating in NEDS waived informed consent because research could not practicably be conducted otherwise. We followed the STROBE reporting guideline. We provided descriptive statistics for patient, visit, and hospital characteristics and calculated 2 JCD-10-based measures of injury severity: Abbreviated Injury Scale (AIS), which grades injuries for each body region from 1 (minor) to 5 (critical),3 and New Injury Severity Score (NISS), which is the sum of squared AIS scores for 3 most severe injuries, ranging from 1 to 75.4 We evaluated injury severity by race and ethnicity and median household income for the home zip code. Differences were assessed using 2-sided weighted Mann-Whitney tests. Analyses incorporated NEDS discharge weights, with P = .05 indicating significance. Analyses were conducted from June 2022 to December 2023, using Microsoft R Open 4.0.2 (Microsoft Corp) (eMethods in Supplement 1). Results | We identified 1276 ED visits with Y35.83X codes (5152 visits after weighting). Patients included 1186 males (92.9%) and 91 females (7.1%) (mean [SD] age, 32.9 [10.4] years) residing in zip codes below the 50th percentile for median household income (67.5%). They presented to teaching hospitals (70.8%) in metropolitan areas (86.1%) and had Asian or Pacific Islander (1.4%), Black (35.7%), Hispanic (17.6%), Native American (1.8%), White (39.2%), or other (4.3%) race and ethnicity. Among patients with injuries (70.8%), 61.1% had minor, 25.4% moderate, 9.0% serious, 2.9% severe, and 1.6% critical injuries (Table 1). Most common injured body regions were the extremities (36.9%), followed by chest (27.4%), head and neck (25.6%), abdomen (24.2%), and face (8.7%). Percentage of serious, severe, or critical injuries ranged from 14.9% for Native American patients to 7.3% for Hispanic patients (Table 2). However, differences were not significant nor were there significant race-based differences in NISS or maximum AIS score. Compared with patients in the upper 2 quartiles of household income, patients in the bottom 2 quartiles experienced jamainternalmedicine.com JAMA Internal Medicine April 2024 Volume 184, Number 4 © 2024 American Medical Association. All rights reserved. Letters Table 2. Injury Severity by Patient Race and Household Income Injury severity Maximum AIS score* NISSa % Of patients with serious, severe, or critical injuries* Variable Weighted mean (SD)b P valuec Weighted mean (SD)b P valuec Weighted Pvaluec Race and ethnicity" Asian or Pacific Islander 0.95 (0.88) 1.73 (2.48) 7.3 Black 1.09(1.00) 2.63 (4.38) 8.2 Hispanic 1.11 (1.03) 63 2.68(5.00) 46 7.3 43 Native American 1.39 (1.24) 3.90(5.50) 14.9 White 1.19(1.06) 2.99(4.39) 11.8 Other6 1.05 (1.03) 2.39(4.12) 10.2 Median household income for patient's home zip code 0-25th Percentile 1.16(1.05) 3.00(5.02) 10.6 26th-50th Percentile 1.21 (1.04) 2.88(4.34) m 9.5 51st-75th Percentile 1.08(1.00) .\JD 2.48(3.72) 9.1 76th-100th Percentile 0.99(1.01) 2.31(4.03) 7.9 Abbreviations: AIS, Abbreviated Injury Scale (score range: 1-5, with the highest score indicating critical injuries); NISS, New Injury Severity Score (range: 1-41, with the highest score indicating multiple severe or critical injuries). a The AIS score was calculated separately for each body region; these values represent the maximum score on each patient record. The NISS is an overall measure of injury severity. Serious, severe, or critical injuries were defined as a maximum AIS score of 3 or higher. b Authors' analysis of Nationwide Emergency Department Sample (NEDS). Weighted means and SDsfor injury severity accounted for the NEDS discharge weights. c Differences between groups were assessed using 2-sided weighted Mann-Whitney tests. dRace and ethnicity data were obtained from NEDS. e No other information was available for this category. higher NISS (Xi = 2.28; P = .02) and maximum AIS score (X? = 2.18;P = .03). Patients with lower income were more likely to experience serious, severe, or critical injuries, but these differences were not significant. Discussion | Most ED visits for CED injuries involved young Black and White males from low-income areas. Black individuals were overrepresented in the sample vs the US population, consistent with research demonstrating increased risk of police violence in Black populations.5 Study limitations include injury totals representing a lower bound, because individuals with minor injuries may not visit EDs. We also could not distinguish CED deployment-related injuries from other injuries before or after arrest. Data were deidentified; thus, we could not identify visits from the same individuals. The study may be underpowered to detect differences in injury patterns between populations. Patients experienced puncture wounds or foreign-body injuries from barb placement and concussions, fractures, or traumatic brain injuries from muscle contractions and falls associated with CED. Police departments should provide adequate CED training to prevent long-term injury and prioritize de-escalation techniques.6 Emma M. Achola, BA Kevin N. Griffith, PhD Jesse O. Wrenn, MD, PhD Carmen R. Mitchell, MPH Dawn Schwartz, MA Christianne L. Roumie, MD, MPH Author Affiliations: Department of Health Policy, Vanderbilt University Medical Center, Nashville, Tennessee (Achola, Griffith, Schwartz, Roumie); Partnered Evidence-Based Policy Resource Center, Veterans Affairs (VA) Boston Healthcare System, Boston, Massachusetts (Griffith); Department of Emergency Medicine, Vanderbilt University Medical Center, Nashville, Tennessee (Wrenn); School of Public Health and Information Sciences, University of Louisville, Louisville, Kentucky (Mitchell); VA Tennessee Valley Health Care System, Geriatric Research Education Clinical Center, Nashville (Roumie); Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee (Roumie). Accepted for Publication: November 17,2023. Published Online: February 5,2024. doiTO.IOOI/jamaintemmed.2023.8012 Corresponding Author: Kevin N. Griffith, PhD, Department of Health Law, Policy & Management, Vanderbilt University Medical Center, 2525 West End Ave, Ste 1204, Nashville, TN 37203 (kevin.griffith@vumc.org). Author Contributions: Ms Achola and Dr Griffith had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptonddesign: Griffith, Wrenn, Mitchell. Acquisition, analysis, or interpretation of data: Achola, Griffith, Wrenn, Schwartz, Roumie. Drafting of the manuscript: Achola, Griffith, Mitchell. Critical review of the manuscript for important intellectual content: Achola, Griffith, Wrenn, Schwartz, Roumie. Statistical analysis: Achola, Griffith. Administrative, technical, or material support: Wrenn, Schwartz, Roumie. Supervision: Griffith, Roumie. Conflict of Interest Disclosures: None reported. Data Sharing Statement: See Supplement 2. 1. Government Accountability Office. Taser weapons: use of tasers by selected law enforcement agencies. 2005. Accessed August 2,2023. https://www.gao. gov/assets/gao-05-464.pdf 2. Owens PL, Barrett ML, Gibson TB, Andrews RM, Weinick RM, Mutter RL. Emergency department care inthe United States: a profile of national data sources. Ann Emerg Med. 2010;56(2):150-165. doi:10.1016/j.annemergmed. 2009.11.022 3. Baker SP, O'Neill B, Haddon W Jr, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974;14(3):187-196. doi:10.1097/00 005373-197403000-00001 442 JAMA Internal Medicine April 2024 Volume 184, Number4 jamainternalmedicine.com © 2024 American Medical Association. All rights reserved. Letters 4. Stevenson M, Segui-Gomez M, Lescohier I, Di Scala C, McDonald-Smith G. An overview of the Injury Severity Score and the new Injury Severity Score. InjPrev. 2001;7(1):10-13. doi:10.1136/ip.7.1.10 5. Motley RO Jr, Joe S. Police use of force by ethnicity, sex, and socioeconomic class. JSocSocial Work Res. 2018;9(1):49-67. doi:10.1086/696355 6. TASER International Inc. TASERX3TM, X26TM, and M26TM ECD warnings, instructions, and information: law enforcement. 2010. Accessed March 6,2023. https://fingfx.thomsonreuters.com/gfx/rngs/USA-TASER/0100503907S/ images/warnings-2010.pdf Emergency Department Use Disparities Among Transgender and Cisgender Medicare Beneficiaries. 2011-2020 Transgender and gender-diverse (TGD) people face substantial societal stigma1 due to their identities in health care settings. TGD individuals often postpone routine medical care due to various reasons, including anticipated discrimination, Editor's Note page 445 lack of knowledgeable clini- cians, and costs. These de- Supplemental content lays are associated with medical emergencies and poor long-term health outcomes.2 We examined national emergency department (ED) use among TGD beneficiaries and explored whether TGD beneficiaries use the ED differently than cisgender beneficiaries. Methods | We analyzed 2011 to 2020 data from a random 20% sample of the Medicare inpatient, carrier, and enrollment files. We used a claims-based algorithm to identify TGD beneficiaries (eMethods in Supplement l).3 Claims-based TGD identification algorithms have high sensitivity and specificity.4 We used a 50% random sample of non-TGD beneficiaries with at least 1 claim as the cisgender comparison group. We identified ED visits using Healthcare Common Procedure Coding System codes or inpatient claims with an ED charge amount of more than $0. We categorized ED visit severity and reason using an established algorithm.5,6 We fit a logistic regression using a generalized estimating equation to predict any ED use, any ED in each of the utilization categories, and any inpatient admission from the ED adjusting for age, race and ethnicity, area deprivation index, dual eligibility, disability, chronic conditions, and months enrolled. We used inverse probability weights to account for observable differences between TGD and cisgender beneficiaries. We conducted additional stratified analyses by the original basis of eligibility, given the imbalance of eligibility pathway for TGD and cisgender beneficiaries. The analytical file was prepared using SAS (SAS Institute), and the analyses were performed in Stata (StataCorp). This study followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting and was deemed exempt by the institutional review board at Brown Table 1. Comparison of Demographic Characteristics Among TGD and Cisgender Beneficiaries as Stratified by Basis of Eligibility All Aged (>65 y) With disabilities Characteristic Cisgender (n = 6151389) TGD (n = 3693) SMD Cisgender (n = 4 701293) TGD (n = 1163) SMD Cisgender (n = 1 396 300) TGD (n = 2474) SMD Age, mean (SD), y 66.7 (12.1) 49.0(17.6) 1.18 70.7 (8.3) 67.5 (5.8) 0.45 53.9(13.5) 40.3 (14.3) 0.98 Race and ethnicity, % Asian 2.8 1.5 0.04 3.2 1.7 0.1 1.5 1.4 0.13 Black 10.1 12.5 NA 7.5 5.3 NA 18.0 15.2 NA Hispanic 6.9 6.2 NA 6.1 5.3 NA 9.3 6.6 NA North American Native 0.5 0.8 NA 0.4 0.3 NA 0.9 1.0 NA Other 0.9 0.8 NA 0.9 0.5 NA 0.9 0.9 NA White 77.2 75.1 NA 80.2 85.0 NA 68.6 71.3 NA Missing 1.6 3.1 NA 1.8 1.9 NA 1.0 3.6 NA Dual status, % NA NA -0.57 NA NA -0.04 NA NA -0.39 Nondual 83.5 58.4 NA 90.0 88.8 NA 62.2 44.1 NA Partial dual 3.5 7.0 A 2.1 2.5 NA 8.3 8.9 NA Fulldual 12.0 33.0 NA 7.2 7.9 NA 27.8 45.0 NA Missing 1.0 1.7 NA 0.7 0.8 NA 1.7 2.1 NA Original basis of eligibility, % NA NA -0.97 NA NA NA NA NA NA Age 76.4 31.5 NA 100 100 NA 0 0 NA Disability 22.7 67.0 NA 0 0 NA 100 100 NA ESKD 0.9 1.5 NA 0 0 NA 0 0 NA Current basis of eligibility NA NA -1.02 NA NA 0.02 NA NA -0.44 Age 80.2 33.9 NA 100 100 NA 16.4 3.4 NA Disability 19.0 64.7 NA 0 0 NA 83.6 96.5 NA ESKD, % 0.8 1.4 NA 0 0 NA 0 0 NA Chronic disease, % NA NA 0.18 NA NA 0.07 NA NA 0.2 0-1 Chronic diseases 46.8 54.5 NA 46.4 50.9 NA 49.3 57.0 NA 2-3 Chronic diseases 21.4 21.2 NA 21.4 18.9 NA 21.2 22.4 NA >4 Chronic diseases 31.8 24.3 NA 32.2 30.2 NA 29.5 20.6 NA Months enrolled, mean (SD) 82.9 (39.3) 87.6(38.2) -0.12 81.4(39.3) 80.4 (37.9) 0.03 88.4 (38.8) 91.3 (37.8) -0.08 ADI score 49.1 (27.3) 50.5 (27.1) -0.05 46.3 (27.0) 44.5 (26.7) 0.07 58.2 (26.5) 53.2 (26.9) 0.19 Abbreviations: ADI, area deprivation index; ESKD, end-stage kidney disease; NA, not applicable; SMD, standard mean difference; TGD, transgender and gender diverse. jamainternalmedicine.com JAMA Internal Medicine April 2024 Volume 184, Number 4 443 © 2024 American Medical Association. All rights reserved. Copyright of JAMA Internal Medicine is the property of American Medical Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.