Samuel Golenbock
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Samuel Golenbock has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Westyn Branch-Elliman
Greater Los Angeles VA Healthcare System
6 shared publications - Hillary J. Mull
VA Boston Healthcare System
5 shared publications - Judith Strymish
Center for Healthcare Organization and Implementation Research (CHOIR), Boston Campus, VA Boston Healthcare System, Boston, MA, USA
3 shared publications - Marlena H. Shin
VA Boston Healthcare System
3 shared publications - Rebecca P. Lamkin
Center for Healthcare Organization and Implementation Research (CHOIR), VA Boston Healthcare System, Boston, MA, USA
3 shared publications Dipandita Basnet
VA Boston Healthcare System
2 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 6 (1) , e358
The accuracy and utility of large language models for extracting stewardship data from handwritten and scanned-in records
Background: Some clinical documentations, including surgical antibiotic administration data and hand-written notes, are only captured in non-digitized scanned documents, limiting their accessibility. Commercially available artificial intelligence (AI) tools may offer a means of extracting such data from scanned-in images, however, little is known about their accuracy for this application. Thus, the objective of this study was to evaluate the performance of two commercially available, protected AI tools for extracting antibiotic administration data as part of a larger study evaluating the sustainability of stewardship goals achieved during the Surgical Care Improvement Project mandatory reporting period. Methods: A sample of surgical cases with antibiotic administration data documented only in non-digitized scanned records was manually reviewed and submitted to two commercially available AI tools (Claude Sonnet version 4.6 and OpenAI version 5.0). The AI tools were prompted to extract data about antibiotic name, dose, and administration time. Results: The total accuracy across all three metrics (drug, dose, timing of administration) was limited (3/14, 21.4% for Claude and 2/14, 14.3% for OpenAI), but overall excellent for identifying that an antibiotic had been administered and the type of drug administered (Claude, 100%, OpenAI, 92.8%). AI tools struggled to identify total dose administered (Claude, 50%, OpenAI, 42.9%), primarily due to challenges distinguishing aliquots from administered doses. Accuracy for identifying timing was also variable, particularly when administration time was documented using a graphic scale rather than written text. Conclusions: Potential utility of these commercial tools for extracting antimicrobial use data from scanned-in records depends upon review goals and what level of precision is necessary.
- 6 November 2026
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 6 (1) , e358
Open AccessThe accuracy and utility of large language models for extracting stewardship data from handwritten and scanned-in records
Background: Some clinical documentations, including surgical antibiotic administration data and hand-written notes, are only captured in non-digitized scanned documents, limiting their accessibility. Commercially available artificial intelligence (AI) tools may offer a means of extracting such data from scanned-in images, however, little is known about their accuracy for this application. Thus, the objective of this study was to evaluate the performance of two commercially available, protected AI tools for extracting antibiotic administration data as part of a larger study evaluating the sustainability of stewardship goals achieved during the Surgical Care Improvement Project mandatory reporting period. Methods: A sample of surgical cases with antibiotic administration data documented only in non-digitized scanned records was manually reviewed and submitted to two commercially available AI tools (Claude Sonnet version 4.6 and OpenAI version 5.0). The AI tools were prompted to extract data about antibiotic name, dose, and administration time. Results: The total accuracy across all three metrics (drug, dose, timing of administration) was limited (3/14, 21.4% for Claude and 2/14, 14.3% for OpenAI), but overall excellent for identifying that an antibiotic had been administered and the type of drug administered (Claude, 100%, OpenAI, 92.8%). AI tools struggled to identify total dose administered (Claude, 50%, OpenAI, 42.9%), primarily due to challenges distinguishing aliquots from administered doses. Accuracy for identifying timing was also variable, particularly when administration time was documented using a graphic scale rather than written text. Conclusions: Potential utility of these commercial tools for extracting antimicrobial use data from scanned-in records depends upon review goals and what level of precision is necessary.
- 7 October 2026
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- p. 1-10
Electronic measurement of compliance with a Surgical Care Improvement Program (SCIP) antimicrobial prophylaxis guidelines in the Veterans health administration
Objective: The Surgical Care Improvement Project (SCIP) was developed and implemented nationally from 2005 to 2015 partly to improve surgical antimicrobial prophylaxis processes. SCIP’s manual review process was costly yet led to high compliance rates in the Veterans Administration (VA). Our objective was to develop an electronic compliance algorithm to replicate the SCIP metric for timely discontinuation of antimicrobial prophylaxis (INF-3) in cardiac surgeries. Study design: This retrospective cohort study matched manually adjudicated INF-3 data from the VA External Peer Review Program (EPRP) with 2006–2015 cardiac surgeries in the Corporate Data Warehouse (CDW). Algorithm development used EPRP data as the gold standard, supplemented with manual review, to apply SCIP-defined electronic exclusions and flag cases with appropriate antimicrobial discontinuation. Algorithm validation used manual review of compliant (n = 90) and non-compliant cases (n = 210) among cardiac surgeries not in EPRP. Results: The algorithm utilized CDW prophylactic versus therapeutic antimicrobial drug names and timestamps for drug administration to reliably exclude cases with appropriate antimicrobial use before flagging INF-3 compliance/non-compliance. In the EPRP-CDW development sample (n = 29,431), the algorithm-measured INF-3 compliance rate was 96.2% versus EPRP’s rate of 95.9%. In validation (n = 10,131), sensitivity was 92.1% (95%CI: 88.0%, 96.2%) and specificity was 97.0% (95%CI: 93.6%, 100%) compared to manual review. The algorithm exclusion rate in CDW cardiac surgeries was 18.5%—primarily for antimicrobial pharmacy administration indicative of pre-or postoperative infection (7,360/52,310 14.1%). Conclusions: The SCIP INF-3 cardiac electronic algorithm effectively replicated the EPRP manual review process and has the potential to extend surveillance of appropriate surgical antimicrobial prophylaxis.
- 7 October 2026
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- p. 1-10
Electronic measurement of compliance with a Surgical Care Improvement Program (SCIP) antimicrobial prophylaxis guidelines in the Veterans health administration
Objective: The Surgical Care Improvement Project (SCIP) was developed and implemented nationally from 2005 to 2015 partly to improve surgical antimicrobial prophylaxis processes. SCIP’s manual review process was costly yet led to high compliance rates in the Veterans Administration (VA). Our objective was to develop an electronic compliance algorithm to replicate the SCIP metric for timely discontinuation of antimicrobial prophylaxis (INF-3) in cardiac surgeries. Study design: This retrospective cohort study matched manually adjudicated INF-3 data from the VA External Peer Review Program (EPRP) with 2006–2015 cardiac surgeries in the Corporate Data Warehouse (CDW). Algorithm development used EPRP data as the gold standard, supplemented with manual review, to apply SCIP-defined electronic exclusions and flag cases with appropriate antimicrobial discontinuation. Algorithm validation used manual review of compliant (n = 90) and non-compliant cases (n = 210) among cardiac surgeries not in EPRP. Results: The algorithm utilized CDW prophylactic versus therapeutic antimicrobial drug names and timestamps for drug administration to reliably exclude cases with appropriate antimicrobial use before flagging INF-3 compliance/non-compliance. In the EPRP-CDW development sample (n = 29,431), the algorithm-measured INF-3 compliance rate was 96.2% versus EPRP’s rate of 95.9%. In validation (n = 10,131), sensitivity was 92.1% (95%CI: 88.0%, 96.2%) and specificity was 97.0% (95%CI: 93.6%, 100%) compared to manual review. The algorithm exclusion rate in CDW cardiac surgeries was 18.5%—primarily for antimicrobial pharmacy administration indicative of pre-or postoperative infection (7,360/52,310 14.1%). Conclusions: The SCIP INF-3 cardiac electronic algorithm effectively replicated the EPRP manual review process and has the potential to extend surveillance of appropriate surgical antimicrobial prophylaxis.
- Open Access
- 23 January 2026
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 47 (4) , 351-356
Leveraging a large language model to support expansion of surveillance activities to include cardiovascular implantable device infections in a large, integrated national healthcare system
Background: Surveillance activities are emerging as exemplar use cases for large language models (LLMs) in health care. The aim of this study was to evaluate the potential for LLMs to support the expansion of surveillance activities to include cardiovascular implantable electronic device (CIED) procedures. Methods: A validated machine learning-based infection flagging tool was applied to a cohort of VA CIED procedures from 7/1/2021 to 9/30/2023; cases with ≥10% probability of CIED infection underwent manual review. Then, a weighted random sample of 50 infected and 50 uninfected cases was reviewed with generative artificial intelligence (GenAI) assistance. GenAI prompts were iteratively refined to extract and classify all components of infection-related variables from clinical notes. Data extracted by GenAI were compared with manual chart reviews to assess infection status and extraction consistency. Results: Among 12,927 CIED procedures, 334 (2.58%) had ≥10% probability of CIED infection. Among 100 sampled cases, 50 of 50 uninfected cases were correctly categorized. Among 50 infection cases, GenAI identified all CIED infections, but the timing of events and the attribution to a preceding procedure were incorrect in 7 of 50 cases. The overall specificity of the GenAI-assisted process was 100% and the sensitivity for accurately classifying timing and attribution of CIED infection events was 82%. Errors in timing improved with iterative prompt updates. Manual chart reviews averaged 25 minutes per chart; the GenAI-assisted process averaged 5–7 minutes per chart. Conclusions: LLMs can help streamline the review process for healthcare-associated infection surveillance, but manual adjudication of output is needed to ensure the correct timeline of events and attribution.
- 23 January 2026
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 47 (4) , 351-356
Open AccessLeveraging a large language model to support expansion of surveillance activities to include cardiovascular implantable device infections in a large, integrated national healthcare system
Background: Surveillance activities are emerging as exemplar use cases for large language models (LLMs) in health care. The aim of this study was to evaluate the potential for LLMs to support the expansion of surveillance activities to include cardiovascular implantable electronic device (CIED) procedures. Methods: A validated machine learning-based infection flagging tool was applied to a cohort of VA CIED procedures from 7/1/2021 to 9/30/2023; cases with ≥10% probability of CIED infection underwent manual review. Then, a weighted random sample of 50 infected and 50 uninfected cases was reviewed with generative artificial intelligence (GenAI) assistance. GenAI prompts were iteratively refined to extract and classify all components of infection-related variables from clinical notes. Data extracted by GenAI were compared with manual chart reviews to assess infection status and extraction consistency. Results: Among 12,927 CIED procedures, 334 (2.58%) had ≥10% probability of CIED infection. Among 100 sampled cases, 50 of 50 uninfected cases were correctly categorized. Among 50 infection cases, GenAI identified all CIED infections, but the timing of events and the attribution to a preceding procedure were incorrect in 7 of 50 cases. The overall specificity of the GenAI-assisted process was 100% and the sensitivity for accurately classifying timing and attribution of CIED infection events was 82%. Errors in timing improved with iterative prompt updates. Manual chart reviews averaged 25 minutes per chart; the GenAI-assisted process averaged 5–7 minutes per chart. Conclusions: LLMs can help streamline the review process for healthcare-associated infection surveillance, but manual adjudication of output is needed to ensure the correct timeline of events and attribution.
- Open Access
- 23 October 2025
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 5 (1) , e279
Development of a feasible and portable electronic flag for near-real-time identification of renal replacement therapy in the Veterans Health Administration
Background: Chronic kidney disease (CKD) is prevalent among US Veterans. Identifying patients undergoing dialysis in real-time is crucial for implementing patient safety measures, including stewardship interventions, such as medication dosing adjustments. Limited feasible and accurate tools exist for near-real-time identification. This study aimed to develop a renal replacement therapy (RRT) flag using structured data in the Veterans Health Administration (VHA) electronic health record (EHR). Methods: Data from Veterans who underwent cardiovascular implantable electronic device (CIED) procedures (9/2015–12/2019) were linked to US Renal Data Systems (USRDS) data. Potential identifiers included outpatient hemodialysis procedure records, community care hemodialysis consults, ICD-10 diagnoses, and serum creatinine (SCr) >4 mg/dL. USRDS served as the comparison standard, and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Logistic regression determined the area under the curve (AUC). Results: Among 37,706 CIED procedures on 34,994 Veterans, 967 patients (2.6%) were identified by USRDS as ever receiving RRT (hemodialysis and peritoneal dialysis or transplant), with 520 (1.4%) actively receiving RRT at the time of CIED. The RRT flag, combining ≥4 outpatient procedures in the prior 30 days, ≥1 consult in the prior year, and/or SCr >4 mg/dL, achieved an AUC of 0.976 (95% CI: 0.97–0.98), with high sensitivity (0.96; 95% CI: 0.94–0.97) and specificity (0.99; 95% CI: 0.99–1.00). The PPV was 0.70 (95% CI: 0.67–0.74). Performance was slightly lower when consults were replaced with ICD codes. Conclusions: We developed an accurate electronic flag using structured data to identify active RRT within VHA among Veterans undergoing invasive procedures, supporting patient safety and care adjustments. This flag addresses a crucial patient safety gap and supports expansion of stewardship efforts.
- 23 October 2025
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 5 (1) , e279
Open AccessDevelopment of a feasible and portable electronic flag for near-real-time identification of renal replacement therapy in the Veterans Health Administration
Background: Chronic kidney disease (CKD) is prevalent among US Veterans. Identifying patients undergoing dialysis in real-time is crucial for implementing patient safety measures, including stewardship interventions, such as medication dosing adjustments. Limited feasible and accurate tools exist for near-real-time identification. This study aimed to develop a renal replacement therapy (RRT) flag using structured data in the Veterans Health Administration (VHA) electronic health record (EHR). Methods: Data from Veterans who underwent cardiovascular implantable electronic device (CIED) procedures (9/2015–12/2019) were linked to US Renal Data Systems (USRDS) data. Potential identifiers included outpatient hemodialysis procedure records, community care hemodialysis consults, ICD-10 diagnoses, and serum creatinine (SCr) >4 mg/dL. USRDS served as the comparison standard, and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Logistic regression determined the area under the curve (AUC). Results: Among 37,706 CIED procedures on 34,994 Veterans, 967 patients (2.6%) were identified by USRDS as ever receiving RRT (hemodialysis and peritoneal dialysis or transplant), with 520 (1.4%) actively receiving RRT at the time of CIED. The RRT flag, combining ≥4 outpatient procedures in the prior 30 days, ≥1 consult in the prior year, and/or SCr >4 mg/dL, achieved an AUC of 0.976 (95% CI: 0.97–0.98), with high sensitivity (0.96; 95% CI: 0.94–0.97) and specificity (0.99; 95% CI: 0.99–1.00). The PPV was 0.70 (95% CI: 0.67–0.74). Performance was slightly lower when consults were replaced with ICD codes. Conclusions: We developed an accurate electronic flag using structured data to identify active RRT within VHA among Veterans undergoing invasive procedures, supporting patient safety and care adjustments. This flag addresses a crucial patient safety gap and supports expansion of stewardship efforts.
- Open Access
- 30 October 2024
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 45 (12) , 1468-1471
Operationalizing an adverse event detection surveillance system to support antimicrobial stewardship activities: perceptions and insights from the SHEA research network
A surveillance system for measuring patient-level antimicrobial adverse drug events (ADE) may support stewardship activities, however, design and implementation questions remain. In this national survey, stewardship experts favored simple, laboratory-based ADE definitions although there were tensions between feasibility, ability to identify attribution without chart review, and importance of specific ADE.
- 30 October 2024
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 45 (12) , 1468-1471
Open AccessOperationalizing an adverse event detection surveillance system to support antimicrobial stewardship activities: perceptions and insights from the SHEA research network
A surveillance system for measuring patient-level antimicrobial adverse drug events (ADE) may support stewardship activities, however, design and implementation questions remain. In this national survey, stewardship experts favored simple, laboratory-based ADE definitions although there were tensions between feasibility, ability to identify attribution without chart review, and importance of specific ADE.
- Open Access
- 25 October 2023
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 3 (1) , e187
Factors associated with uptake of guideline-recommended cardiovascular implantable electronic device management: a nationwide, retrospective cohort study
Clinical guidelines recommend device removal for cardiovascular implantable electronic device (CIED) infection management. In this retrospective, nationwide cohort, 60.8% of CIED infections received guideline-concordant care. One-year mortality was higher among those without procedural management (25% vs 16%). Factors associated with receipt of device procedures included pocket infections and positive microbiology.
- 25 October 2023
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 3 (1) , e187
Open AccessFactors associated with uptake of guideline-recommended cardiovascular implantable electronic device management: a nationwide, retrospective cohort study
Clinical guidelines recommend device removal for cardiovascular implantable electronic device (CIED) infection management. In this retrospective, nationwide cohort, 60.8% of CIED infections received guideline-concordant care. One-year mortality was higher among those without procedural management (25% vs 16%). Factors associated with receipt of device procedures included pocket infections and positive microbiology.
- 1 February 2021
- Published by British Editorial Society of Bone & Joint Surgery in The Bone & Joint Journal
- Vol. 103-B (2) , 360-365
Reassessing glenoid inclination in reverse total shoulder arthroplasty with glenosphere lateralization
Aims: Existing literature indicates that inferiorly inclined glenoid baseplates following reverse total shoulder arthroplasty (RSA) produce better outcomes compared to superiorly inclined baseplates. We aim to compare clinical outcomes for RSAs with superiorly and neutrally/inferiorly inclined lateralized glenospheres. Methods: We retrospectively reviewed 154 consecutive patients undergoing RSA between July 2015 and July 2017 by one single-fellowship trained surgeon (AJ). Two raters (KAM and MVS) independently measured glenoid inclination in preoperative and minimum two year follow-up radiographs (anteroposterior/Grashey) using the RSA angle. Inclination was then compared to patient-reported outcomes, range of motion (ROM), and independently assessed degree of scapular notching and staging of heterotopic ossification at two year follow-up. Results: Median postoperative inclination for each group was found to be -3.6° (interquartile range (IQR) -2.1 to -6.9) and 6.0° (3.2° to 10.1°) for the neutrally/inferiorly and superiorly inclined cohorts, respectively. Preoperative inclination was highly associated with postoperative inclination (p = 0.004). When comparing superiorly and neutrally/inferiorly inclined glenospheres, there were no differences in heterotopic ossification (p = 0.606), scapular notching (p = 0.367), American Shoulder and Elbow Surgeons score (p = 0.419), Single Assessment Numeric Evaluation (p = 0.417), Visual Analogue Scale (VAS) pain score (p = 0.290), forward elevation (p = 0.161), external rotation (p = 0.537), or internal rotation (p = 0.656). Conclusion: Compared to neutral and inferior inclination, up to 6° ± 3° of superior glenoid baseplate inclination on a lateralized RSA design produces no differences in postoperative ROM or patient-reported outcomes, and produces similar levels of scapular notching and heterotopic ossification. Additionally, the degree of preoperative inclination represents an important factor in surgical decision-making as it is strongly associated with postoperative inclination. It is important to note that the findings of this study are only reflective of lateralized RSA prostheses. Cite this article: Bone Joint J 2021;103-B(2):360–365.
- 1 February 2021
- Published by British Editorial Society of Bone & Joint Surgery in The Bone & Joint Journal
- Vol. 103-B (2) , 360-365
Reassessing glenoid inclination in reverse total shoulder arthroplasty with glenosphere lateralization
Aims: Existing literature indicates that inferiorly inclined glenoid baseplates following reverse total shoulder arthroplasty (RSA) produce better outcomes compared to superiorly inclined baseplates. We aim to compare clinical outcomes for RSAs with superiorly and neutrally/inferiorly inclined lateralized glenospheres. Methods: We retrospectively reviewed 154 consecutive patients undergoing RSA between July 2015 and July 2017 by one single-fellowship trained surgeon (AJ). Two raters (KAM and MVS) independently measured glenoid inclination in preoperative and minimum two year follow-up radiographs (anteroposterior/Grashey) using the RSA angle. Inclination was then compared to patient-reported outcomes, range of motion (ROM), and independently assessed degree of scapular notching and staging of heterotopic ossification at two year follow-up. Results: Median postoperative inclination for each group was found to be -3.6° (interquartile range (IQR) -2.1 to -6.9) and 6.0° (3.2° to 10.1°) for the neutrally/inferiorly and superiorly inclined cohorts, respectively. Preoperative inclination was highly associated with postoperative inclination (p = 0.004). When comparing superiorly and neutrally/inferiorly inclined glenospheres, there were no differences in heterotopic ossification (p = 0.606), scapular notching (p = 0.367), American Shoulder and Elbow Surgeons score (p = 0.419), Single Assessment Numeric Evaluation (p = 0.417), Visual Analogue Scale (VAS) pain score (p = 0.290), forward elevation (p = 0.161), external rotation (p = 0.537), or internal rotation (p = 0.656). Conclusion: Compared to neutral and inferior inclination, up to 6° ± 3° of superior glenoid baseplate inclination on a lateralized RSA design produces no differences in postoperative ROM or patient-reported outcomes, and produces similar levels of scapular notching and heterotopic ossification. Additionally, the degree of preoperative inclination represents an important factor in surgical decision-making as it is strongly associated with postoperative inclination. It is important to note that the findings of this study are only reflective of lateralized RSA prostheses. Cite this article: Bone Joint J 2021;103-B(2):360–365.
- 14 May 2020
- Published by SAGE Publications in Shoulder & Elbow
- Vol. 13 (5) , 527-533
Reassessing the minimum two-year follow-up standard after total shoulder arthroplasty—Is one year sufficient?
US Food and Drug Administration Investigation Device Exemption studies and academic journals emphasize the importance of two-year follow-up data in reporting outcomes of total shoulder arthroplasty, but there is limited data evaluating appropriate follow-up length. We aim to evaluate change in postoperative outcomes and complications between one and two years following anatomic and reverse total shoulder arthroplasties. We retrospectively identified 250 patients who underwent anatomic and reverse total shoulder arthroplasties between 2013 and 2016 from a single surgeon arthroplasty registry. Patients without both one- and two-year follow-up data were excluded. We compared American Shoulder and Elbow Surgeons (ASES) score, Visual Analog Scale for pain, and goniometer-measured range of motion. Patient-reported outcome measurements (p > 0.05) did not change between one and two years postoperatively following both reverse (n = 146) and anatomic (n = 104) total shoulder arthroplasties. Range of motion increased slightly (p < 0.05), but this change was not clinically relevant. There were no additional complications. Minimum two-year clinical follow-up may not be necessary for future shoulder arthroplasty Investigation Device Exemption and other peer-reviewed investigations. Patient-reported outcomes (ASES and pain score) and range of motion plateau at one year postoperatively without additional complications. One-year follow-up is an acceptable minimum follow-up length. Level III—retrospective analysis.
- 14 May 2020
- Published by SAGE Publications in Shoulder & Elbow
- Vol. 13 (5) , 527-533
Reassessing the minimum two-year follow-up standard after total shoulder arthroplasty—Is one year sufficient?
US Food and Drug Administration Investigation Device Exemption studies and academic journals emphasize the importance of two-year follow-up data in reporting outcomes of total shoulder arthroplasty, but there is limited data evaluating appropriate follow-up length. We aim to evaluate change in postoperative outcomes and complications between one and two years following anatomic and reverse total shoulder arthroplasties. We retrospectively identified 250 patients who underwent anatomic and reverse total shoulder arthroplasties between 2013 and 2016 from a single surgeon arthroplasty registry. Patients without both one- and two-year follow-up data were excluded. We compared American Shoulder and Elbow Surgeons (ASES) score, Visual Analog Scale for pain, and goniometer-measured range of motion. Patient-reported outcome measurements (p > 0.05) did not change between one and two years postoperatively following both reverse (n = 146) and anatomic (n = 104) total shoulder arthroplasties. Range of motion increased slightly (p < 0.05), but this change was not clinically relevant. There were no additional complications. Minimum two-year clinical follow-up may not be necessary for future shoulder arthroplasty Investigation Device Exemption and other peer-reviewed investigations. Patient-reported outcomes (ASES and pain score) and range of motion plateau at one year postoperatively without additional complications. One-year follow-up is an acceptable minimum follow-up length. Level III—retrospective analysis.