Michael George
VerifiedUniversity of Pennsylvania · Rehabilitation Medicine
Active 1901–2025
About
Michael George, MD, MSCE, is an Assistant Professor of Medicine (Rheumatology) at the Hospital of the University of Pennsylvania. He serves as the Program Director for the Rheumatology Fellowship at the University of Pennsylvania. Dr. George completed his undergraduate studies at Harvard University, graduating Magna Cum Laude in 2005. He earned his MD from Johns Hopkins University School of Medicine in 2009 and obtained a Master of Science in Clinical Epidemiology (MSCE) from the University of Pennsylvania in 2016. His professional focus includes rheumatology, with research contributions in the areas of autoimmune diseases, treatment strategies, and health care disruptions during the COVID-19 pandemic.
Research topics
- Pathology
- Intensive care medicine
- Internal medicine
- Medicine
- Physical therapy
- Emergency medicine
- Virology
Selected publications
RMD Open · 2025-07-01 · 1 citations
articleOpen accessSenior authorOBJECTIVES: To assess vaccine effectiveness (VE) and safety of the recombinant zoster vaccine (RZV) in adults with systemic lupus erythematosus (SLE). METHODS: This retrospective study using administrative claims data collected between 1 January 2018 and 30 September 2023 from the USA included two cohorts of adults aged ≥18 years with SLE insured by (1) Medicare (parts A/B/D) or (2) one of five commercial partners (including Medicare Advantage plans). In VE analyses, adults receiving two RZV doses ≥28 days apart were matched 1:4 to unvaccinated comparators on insurer, sex and age (±5 years); the outcome was incident herpes zoster (HZ) after 31 days. In safety analyses, adults receiving RZV dose 1 or 2 were matched to unvaccinated comparators as above; the outcome was severe SLE flare within 90 days. Adjusted hazard ratios (HRs) were estimated using Cox models with inverse probability of treatment weighting (IPTW). RESULTS: VE cohorts included 2284 Medicare and 1308 commercially insured RZV-vaccinated patients; mean weighted follow-up was 1.4-1.6 person-years. Safety cohorts included 6602 RZV vaccinations (dose 1 or 2) in Medicare and 4196 in commercial insurers. Vaccinated versus unvaccinated patient characteristics were balanced after IPTW. VE was 70% (95% CI: 50% to 82%) in Medicare and 54% (95% CI: 18% to 74%) in commercially insured patients. The HR of severe SLE flare in vaccinated versus unvaccinated patients was 0.91 (95% CI: 0.75 to 1.11) in Medicare and 0.94 (95% CI: 0.72 to 1.24) in commercially insured patients. CONCLUSIONS: RZV prevented a majority of HZ cases in individuals with SLE without increasing the risk of severe SLE flare.
ACR Open Rheumatology · 2025-04-01 · 13 citations
articleOpen accessOBJECTIVE: This study aimed to report the safety of avacopan, an oral selective complement C5a receptor antagonist, using pooled data from clinical trials in patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (granulomatosis with polyangiitis [GPA] or microscopic polyangiitis [MPA]). METHODS: Data were included from two phase 2 (CLEAR [NCT01363388] and CLASSIC [NCT02222155]) and one phase 3 (ADVOCATE [NCT02994927]) double-blind randomized controlled trials comparing the safety and efficacy of avacopan with active non-avacopan control regimens to induce remission in patients with GPA or MPA. In CLEAR and ADVOCATE, avacopan-treated patients received either no or lower doses of study-supplied prednisone than the control groups; in CLASSIC, all groups received the same dose of study-supplied prednisone. Assessments included rates of exposure-adjusted adverse events (AEs), serious AEs (SAEs), and AEs of special interest. RESULTS: Overall, 439 patients with GPA or MPA (avacopan: n = 239; non-avacopan: n = 200) were included. The exposure-adjusted rates of AEs, SAEs, white blood cell (WBC) count reductions, and infections were lower with avacopan versus control (between-group differences in rate per 100 patient-years -151.9 [95% confidence interval (CI) -218.6 to -85.3], -20.8 [95% CI -38.3 to -3.3], -11.6 [95% CI -22.2 to -1.2], and -24.3 [95% CI -48.5 to -0.1], respectively). SAEs associated with hepatic function abnormalities occurred in 4.4% of the avacopan group and 2.8% of the control group. CONCLUSION: In clinical trials of GPA or MPA, use of avacopan was associated with fewer AEs, SAEs, and WBC count reductions and fewer infections than non-avacopan treatment. Safety data support the use of avacopan in patients with GPA or MPA.
The Lancet Rheumatology · 2025-01-07 · 14 citations
articleOpen accessGenetically-determined variation in CRP impacts disease activity assessment in rheumatoid arthritis
Lara D. Veeken · 2025-03-18 · 3 citations
articleOpen accessOBJECTIVES: In patients with RA, we evaluated a single-nucleotide variant previously associated with lower CRP to assess if this impacts clinical disease activity assessments including the disease activity score-28 with CRP (DAS28(CRP)). METHODS: Patients from three observational cohorts were evaluated-the United Kingdom Biobank (UKB), the Veterans Affairs RA Registry (VARA) and the FORWARD Databank. The effect of rs1205 genotype on log-adjusted serum CRP concentrations was assessed using linear regression adjusted for sex, age and population structure. The regression coefficients from UKB were converted to create a modified CRP and DAS28(CRP). Reclassification to an abnormal CRP (>0.8 mg/dl) and between DAS28(CRP) disease activity groups were determined. RESULTS: Among the 488 377 UKB participants, individuals with the TT and CT genotype had a statistically significantly lower log-adjusted CRP compared with the CC genotype [β TT genotype: -0.371 (95% CI -0.381, -0.360), P < 0.001; β CT genotype: -0.173 (95% CI -0.179, -0.167), P < 0.001]. In the 2597 VARA participants, the DAS28(CRP) was significantly lower in the genotype TT compared with the CC genotype [β -0.22, (95% CI -0.44, -0.0027), P = 0.047], but not in the CT genotype. In those with the TT genotype, 6-8% were reclassified to having an abnormal CRP and 3% of patients were reclassified from low to moderate disease activity. CONCLUSION: The rs1205 TT genotype in CRP is associated with lower CRP and DAS28(CRP). Given the widespread use of CRP, this demonstrates that genetic factors, irrespective of an individual patient's disease activity, can cause differences in CRP that impact disease activity assessment.
Arthritis & Rheumatology · 2025-01-13
letterOpen accessDisclosure form Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Arthritis Care & Research · 2025-08-19 · 1 citations
articleOpen accessOBJECTIVE: We assessed whether circulating adipokines are associated with incident fractures in patients with rheumatoid arthritis (RA). METHODS: Three adipokines (adiponectin, leptin, and fibroblast growth factor [FGF]-21) were measured using banked enrollment serum from participants in a longitudinal RA cohort. Adipokine levels were dichotomized as high/low using median values. Incident osteoporotic fracture was defined based on published algorithms using diagnostic codes and confirmed by chart review. Cox proportional hazard models evaluated adipokines and incident fracture risk adjusting for age, sex, race, smoking status, body mass index (BMI), prednisone use, disease activity, comorbidity score, calendar year, osteoporosis history, and previous fracture. RESULTS: A total of 2,527 participants were included (89% male, mean age 72 years). There were 228 incident fractures over 27,540 person-years of follow-up (8.3 fractures per 1,000 person-years). After adjustment, the risk of incident fracture was increased for high levels of leptin (hazard ratio [HR] = 1.47; 95% confidence interval [CI] 1.15-1.90; P = 0.003), FGF-21 [HR = 1.39; 95% CI 1.16-1.67; P < 0.001), and adiponectin (HR = 1.21; 95% CI 0.94-1.55), with the latter not achieving significance (P = 0.13). Participants who had elevated levels of all three adipokines experienced twice the risk of fracture compared with those in whom none was elevated (HR = 2.17; 95% CI 1.27-3.70; P = 0.005). CONCLUSION: Elevations in adipokines are associated with an increased risk of fracture in patients with RA, independent of other established risk factors including BMI, smoking, and prednisone use. This supports further investigation to understand whether this association is related to altered body composition or disrupted metabolic pathways.
Identification of Adult Patients With Dermatomyositis Using Real‐World Data Sources
Arthritis Care & Research · 2025-08-12 · 1 citations
articleOpen accessOBJECTIVE: Studying rare diseases like dermatomyositis (DM) in single-center cohorts is challenging due to small sample sizes and limited generalizability. This study develops and evaluates case identification algorithms for DM to enable coordinated analysis across multiple data sources. METHODS: Case identification algorithms were developed to identify adult patients with DM within 11 independent electronic health record or claims databases, totaling over 800 million patients, using the Observational Medical Outcomes Partnership Common Data Model. Algorithm performance was assessed through manual chart review and using Observational Health Data Sciences and Informatics open-source tools (CohortDiagnostics and PheValuator), which quantify incidence rates and performance metrics such as sensitivity and positive predictive value (PPV). RESULTS: Eight DM case identification algorithms were evaluated across 11 databases, revealing significant variability in performance, with sensitivity and PPV differing by more than 30% between some databases. Overall, we identified one incidence algorithm and one prevalence algorithm with good performance, demonstrated by sensitivity rates of 42% and 49% and PPV values of 83% and 84%, respectively. PheValuator quantified algorithm performance within each database, allowing for direct comparison of different criteria. Additionally, CohortDiagnostics generated incidence rates stratified by age decile and sex, aligning with previous epidemiologic data. CONCLUSION: We developed and validated multiple DM case identification algorithms across diverse databases, demonstrating their accuracy through multiple evaluation methods. This approach enables more generalizable, reproducible research using real-world data and can be applied to other rheumatic diseases.
Arthritis Care & Research · 2025-06-02
articleOpen accessOBJECTIVE: Administrative claims are used to evaluate oral glucocorticoid use in rheumatoid arthritis (RA), despite limited evidence to support accuracy. We aimed to evaluate the performance of claims-based algorithms for glucocorticoid use compared to self-report in an RA population. METHODS: Participants with RA enrolled at seven Veterans Affairs Rheumatoid Arthritis (VARA) Registry sites were asked six questions as part of clinical care assessing current prednisone use and dose, recent use, "stockpiling," and receiving prednisone outside the Department of Veterans Affairs (VA). Algorithms using VA prescription claims operationalized current use (active prescription on date of self-report assessment), current dose (that prescription's mean dose), and recent use (active course overlapping the prior 30 or 90 days). We assessed performance characteristics and agreement, benchmarked on self-report. RESULTS: Of 284 participants, 13% reported current prednisone use and 20% reported 90-day use. Sensitivity, specificity, positive predictive value, and negative predictive value were 0.70, 0.98, 0.84, and 0.96, respectively, for current use and 0.71, 0.92, 0.72, and 0.92, respectively, for 90-day use. Cohen's κ was 0.68 for current use and 0.63 for 90-day use. Among participants reporting ≤5 mg/day, agreement for dose was high (weighted κ 0.67). One in four participants reported a stockpile, and one in four reported receiving prednisone from a non-VA provider. CONCLUSION: Algorithms derived from VA claims detecting prednisone prescriptions have high validity compared to patient self-report. The modest sensitivity of these algorithms may reflect stockpiling and non-VA prescriptions. These findings form a basis for contextualizing real-world studies of glucocorticoid use in RA and improve clinical estimation of glucocorticoid use not captured in claims.
Respiratory Medicine · 2025-01-22
articleOpen accessAntinuclear antibodies (ANA) are often found in ILD; whether ANA is associated with radiographic progression of quantitive interstitial lung changes is unknown. We performed longitudinal analyses of adults in the Multi-Ethnic Study of Atherosclerosis using linear mixed effects models with random intercept and slope to evaluate whether baseline ANA was associated with change in the amount of lung with high attenuation areas on CT (HAAs, percentage of imaged lung with -600 to -250 HU). In 6,638 subjects with 17,293 CT scans over 18 years, 741 (11 %) were ANA positive. ANA was not associated with HAA progression with ANA as a dichotomous variable (0.13 % less progression per year for ANA positive vs negative, 95%CI -0.33 %-0.07 %, p = 0.19) or as a continuous variable (0.004 % less progression per year per 10 % increase in ANA, 95%CI -0.01 %-0.005 %, p = 0.37). ANA was not associated with progression of HAA in community dwelling adults.
Seminars in Arthritis and Rheumatism · 2025-04-21 · 8 citations
articleOpen accessOBJECTIVES: This study aimed to compare treatment outcomes in rheumatoid arthritis-associated interstitial lung disease (RA-ILD) between initiators of rituximab, abatacept, tocilizumab, and tofacitinib using the Target Trial Emulation Framework. METHODS: We emulated three trials comparing abatacept, tocilizumab, and tofacitinib with rituximab (reference). Patients fulfilling validated RA-ILD algorithms initiating one of these non-TNFi b/tsDMARDs were propensity score (PS)-matched (1:1) using national Veterans Affairs (VA) data from 2006 to 2020. PS models included demographics, comorbidities, general health status indicators, and several RA- and ILD-related severity measures. Composite study outcomes were death and respiratory-related hospitalization, ascertained by VA data and linkages to the National Death Index and Medicare, over three-year (primary) and one-year follow-up periods (secondary). Cox regression models were used to analyze study outcomes adjusting for any unbalanced variables. Several sensitivity and subgroup analyses were performed. RESULTS: In the primary cohort, we 1:1 matched abatacept (n = 150), tocilizumab (n = 73), and tofacitinib (n = 94) with equal numbers of rituximab initiators (mean age 68.1-69.4 years, 88-92 % male). There were no significant differences in the primary composite outcome among any of the comparisons (abatacept aHR: 1.03 [0.72, 1.47]; tocilizumab aHR: 1.15 [0.68, 1.93]; tofacitinib aHR: 0.89 [0.54, 1.46]). Secondary, subgroup, and sensitivity analyses supported the main findings. CONCLUSIONS: We did not find significant differences in mortality or respiratory hospitalization between RA-ILD patients initiating different non-TNFi b/tsDMARDs, though estimates were imprecise, and residual confounding may be present. These findings emphasize the need for clinical trials of advanced immunomodulatory therapies in RA-ILD.
Recent grants
Frequent coauthors
- 139 shared
Joshua F. Baker
Hospital of the University of Pennsylvania
- 75 shared
Bryant R. England
VA Nebraska Western Iowa Health Care System
- 62 shared
Ted R. Mikuls
Boston University
- 45 shared
Brian C. Sauer
Lake City VA Medical Center
- 44 shared
Jeffrey R. Curtis
University of Alabama at Birmingham
- 44 shared
Spenge Beveridge
St Mary's Hospital
- 44 shared
Smith Percy
Warneford Hospital
- 40 shared
Bolton Shaw
Labs
Michael George LabPI
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Michael George
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup