Rachel McMillan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rachel McMillan has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Gregory Russell Madden
Department of Microbiology, Immunology, and Cancer Biology, School of Medicine University of Virginia Charlottesville Virginia USA
3 shared publications - William A. Petri
Division of Infectious Diseases and International Health, University of Virginia School of Medicine
3 shared publications Emmanuel Lee
University of Virginia School of Medicine, Charlottesville, VA, USA
2 shared publicationsJennie Z. Ma
Department of Public Health Sciences University of Virginia School of Medicine Charlottesville Virginia USA
2 shared publicationsMd. Mostofa Shakil
The International Centre for Diarrhoeal Disease and Research, Bangladesh (Icddr,b) Dhaka Bangladesh
2 shared publicationsCosti D. Sifri
Division of Infectious Diseases & International Health, Department of Medicine, University of Virginia School of Medicine, Charlottesville, VA, USA; Office of Hospital Epidemiology/Infection Prevention & Control, University of Virginia School of Medicine, Charlottesville, VA, USA
1 shared publication
Latest Publications
- Open Access
- 4 November 2026
- Published by Wiley in BJOG: An International Journal of Obstetrics & Gynaecology
First‐Trimester Maternal Gut Microbiome and Small Vulnerable Newborns: A Prospective Cohort Study
Objectives: To investigate the relationship between the first trimester gut microbiome and small vulnerable newborns (SVN). Setting: Rural low‐ and middle‐income country: Bangladesh. Population: 213 women were enrolled during the first trimester of pregnancy as part of the NICHD Global Network for Women's and Children's Health registry, Bangladesh. Methods: Observational study of the maternal gut microbiome and birth outcomes. Blood and stool samples were collected at enrollment. Birth outcomes were recorded within 72 h of delivery. Random forest machine learning models trained on microbiome data were used to predict SVN or first‐trimester maternal features of interest. Hold out sets were used to test model accuracy. Main Outcome Measures: SVN (born preterm, low birth weight, or small for gestational age). Results: The birth outcome SVN ( n = 106) was discriminated from non‐SVN ( n = 107) using a microbiome‐trained random forest model (AUC = 0.653, 95% CI 0.518–0.789). The microbiome also predicted the inflammatory markers plasma alpha‐1‐acid glycoprotein and faecal calprotectin (AUC = 0.742, 95% CI: 0.616–0.868 and AUC = 0.793, 95% CI: 0.680–0.905, respectively), while micronutrient concentrations, anaemia, C‐reactive protein and the presence of atypical Enteropathogenic E. coli were not reliably predicted. Higher alpha‐1‐acid glycoprotein was significantly associated with SVN (aOR = 1.98; 95% CI: 1.04–3.85) whereas calprotectin was not. Conclusions: In this prospective cohort, first‐trimester maternal gut microbiome composition showed modest discrimination for SVN and a marker of systemic inflammation. These findings support further investigation of early pregnancy microbiome–inflammation pathways in fetal growth restriction and preterm birth.
- 4 November 2026
- Published by Wiley in BJOG: An International Journal of Obstetrics & Gynaecology
Open AccessFirst‐Trimester Maternal Gut Microbiome and Small Vulnerable Newborns: A Prospective Cohort Study
Objectives: To investigate the relationship between the first trimester gut microbiome and small vulnerable newborns (SVN). Setting: Rural low‐ and middle‐income country: Bangladesh. Population: 213 women were enrolled during the first trimester of pregnancy as part of the NICHD Global Network for Women's and Children's Health registry, Bangladesh. Methods: Observational study of the maternal gut microbiome and birth outcomes. Blood and stool samples were collected at enrollment. Birth outcomes were recorded within 72 h of delivery. Random forest machine learning models trained on microbiome data were used to predict SVN or first‐trimester maternal features of interest. Hold out sets were used to test model accuracy. Main Outcome Measures: SVN (born preterm, low birth weight, or small for gestational age). Results: The birth outcome SVN ( n = 106) was discriminated from non‐SVN ( n = 107) using a microbiome‐trained random forest model (AUC = 0.653, 95% CI 0.518–0.789). The microbiome also predicted the inflammatory markers plasma alpha‐1‐acid glycoprotein and faecal calprotectin (AUC = 0.742, 95% CI: 0.616–0.868 and AUC = 0.793, 95% CI: 0.680–0.905, respectively), while micronutrient concentrations, anaemia, C‐reactive protein and the presence of atypical Enteropathogenic E. coli were not reliably predicted. Higher alpha‐1‐acid glycoprotein was significantly associated with SVN (aOR = 1.98; 95% CI: 1.04–3.85) whereas calprotectin was not. Conclusions: In this prospective cohort, first‐trimester maternal gut microbiome composition showed modest discrimination for SVN and a marker of systemic inflammation. These findings support further investigation of early pregnancy microbiome–inflammation pathways in fetal growth restriction and preterm birth.
- Open Access
Antibiotics promote susceptibility to C. difficile infection through a CCR5-dependent immune response
Antibiotics (ABXs) represent the current standard of care for treating Clostridioides difficile infection (CDI). Paradoxically, ABX-induced dysbiosis is the primary risk factor for CDI, as disruption of the colonic microbial ecosystem creates an opportunity for C. difficile colonization. Given that ABXs can also alter immune responses, we investigated whether ABXs prime the colonic immune milieu for CDI susceptibility. Here, we implicate ABXs in driving CDI severity through the emergence of pathogenic CCR5-reliant immune populations in the mouse colon. High-throughput immune cell profiling revealed that ABXs shift the colonic immune compartment toward a CCR5-associated type I immunity signature, marked by an expansion of CCR5 + ILC1s and CCR5 + Th1 cells. A partial genetic deletion of CCR5 reversed CDI severity, alleviating colonic inflammation and improving survival. Pharmacological inhibition of the CCL3/4/5-CCR5 circuit also recapitulated these favorable disease outcomes, which we attribute to reduced colonic CCR5 + ILC1, CCR5 + Th1, and CCR5 + CD8 T cell populations during CDI. Together, our findings extend beyond dysbiosis as the canonical CDI risk factor and establish ABX-induced immune imbalance as an underappreciated determinant of CDI susceptibility.
Open AccessAntibiotics promote susceptibility to C. difficile infection through a CCR5-dependent immune response
Antibiotics (ABXs) represent the current standard of care for treating Clostridioides difficile infection (CDI). Paradoxically, ABX-induced dysbiosis is the primary risk factor for CDI, as disruption of the colonic microbial ecosystem creates an opportunity for C. difficile colonization. Given that ABXs can also alter immune responses, we investigated whether ABXs prime the colonic immune milieu for CDI susceptibility. Here, we implicate ABXs in driving CDI severity through the emergence of pathogenic CCR5-reliant immune populations in the mouse colon. High-throughput immune cell profiling revealed that ABXs shift the colonic immune compartment toward a CCR5-associated type I immunity signature, marked by an expansion of CCR5 + ILC1s and CCR5 + Th1 cells. A partial genetic deletion of CCR5 reversed CDI severity, alleviating colonic inflammation and improving survival. Pharmacological inhibition of the CCL3/4/5-CCR5 circuit also recapitulated these favorable disease outcomes, which we attribute to reduced colonic CCR5 + ILC1, CCR5 + Th1, and CCR5 + CD8 T cell populations during CDI. Together, our findings extend beyond dysbiosis as the canonical CDI risk factor and establish ABX-induced immune imbalance as an underappreciated determinant of CDI susceptibility.
- Open Access
- 22 December 2025
- Published by Wiley
The first-trimester maternal gut microbiome is a predictor of the birth of small vulnerable newborns in rural Bangladesh: A prospective cohort study
Objectives: To investigate the relationship between the first trimester microbiome, inflammation, and small vulnerable newborns (SVN). Setting: Rural low, middle-income country: Bangladesh Population: 213 pregnant women were enrolled during the first trimester of pregnancy as part of the NICHD Global Network for Women’s and Children’s Health registry, Bangladesh. Methods: Blood and stool samples were collected at enrollment and birth outcomes within 72 hours of delivery. Using random forest machine learning models, whether the first-trimester maternal microbiome predicted SVN was evaluated. Main outcome measures: Small vulnerable newborns (born preterm, low birth weight, or small for gestational age). Results: The birth outcome SVN could be discriminated from non-SVN using a microbiome-trained random forest model (AUC = 0.653, 95% CI 0.518-0.789) supporting that the microbiome can influence gestation and fetal development. The microbiome also predicted the inflammatory markers plasma alpha-1-acid glycoprotein and fecal calprotectin (AUC = 0.742, 95% CI: 0.616-0.868 and AUC = 0.793, 95% CI: 0.680-0.905, respectively), while micronutrient concentrations, anemia, C-reactive protein, and the presence of atypical Enteropathogenic E. coli were not reliably predicted. Connecting inflammation to SVN, it was found that higher α-1-acid glycoprotein was significantly associated with SVN (logistic regression: aOR = 1.98; 95% CI: 1.04-3.85) whereas no significant association between calprotectin and SVN was observed. Conclusions: Together, these findings support that the first-trimester maternal gut microbiome contributes to SVN via induction of chronic systemic inflammation but not micronutrient absorption and demonstrates that the pathogenesis of SVN starts as early as the first trimester.
- 22 December 2025
- Published by Wiley
Open AccessThe first-trimester maternal gut microbiome is a predictor of the birth of small vulnerable newborns in rural Bangladesh: A prospective cohort study
Objectives: To investigate the relationship between the first trimester microbiome, inflammation, and small vulnerable newborns (SVN). Setting: Rural low, middle-income country: Bangladesh Population: 213 pregnant women were enrolled during the first trimester of pregnancy as part of the NICHD Global Network for Women’s and Children’s Health registry, Bangladesh. Methods: Blood and stool samples were collected at enrollment and birth outcomes within 72 hours of delivery. Using random forest machine learning models, whether the first-trimester maternal microbiome predicted SVN was evaluated. Main outcome measures: Small vulnerable newborns (born preterm, low birth weight, or small for gestational age). Results: The birth outcome SVN could be discriminated from non-SVN using a microbiome-trained random forest model (AUC = 0.653, 95% CI 0.518-0.789) supporting that the microbiome can influence gestation and fetal development. The microbiome also predicted the inflammatory markers plasma alpha-1-acid glycoprotein and fecal calprotectin (AUC = 0.742, 95% CI: 0.616-0.868 and AUC = 0.793, 95% CI: 0.680-0.905, respectively), while micronutrient concentrations, anemia, C-reactive protein, and the presence of atypical Enteropathogenic E. coli were not reliably predicted. Connecting inflammation to SVN, it was found that higher α-1-acid glycoprotein was significantly associated with SVN (logistic regression: aOR = 1.98; 95% CI: 1.04-3.85) whereas no significant association between calprotectin and SVN was observed. Conclusions: Together, these findings support that the first-trimester maternal gut microbiome contributes to SVN via induction of chronic systemic inflammation but not micronutrient absorption and demonstrates that the pathogenesis of SVN starts as early as the first trimester.
- Open Access
- 17 July 2024
- Published by Elsevier in EBioMedicine
- Vol. 106, 105244
Predicting Clostridioides difficile infection outcomes with explainable machine learning
- 17 July 2024
- Published by Elsevier in EBioMedicine
- Vol. 106, 105244
Open AccessPredicting Clostridioides difficile infection outcomes with explainable machine learning
- Open Access
- 9 May 2024
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 45 (9) , 1064-1072
Validation of clinical risk tools for recurrent Clostridioides difficile infection
Objective: We sought to validate available tools for predicting recurrent C. difficile infection (CDI) including recurrence risk scores (by Larrainzar-Coghen, Reveles, D’Agostino, Cobo, and Eyre et al) alongside consensus guidelines risk criteria, the leading severity score (ATLAS), and PCR cycle threshold (as marker of fecal organism burden) using electronic medical records.Design: Retrospective cohort study validating previously described tools.Setting: Tertiary care academic hospital.Patients: Hospitalized adult patients with CDI at University of Virginia Medical Center.Methods: Risk scores were calculated within ±48 hours of index CDI diagnosis using a large retrospective cohort of 1,519 inpatient infections spanning 7 years and compared using area under the receiver operating characteristic curve (AUROC) and the DeLong test. Recurrent CDI events (defined as a repeat positive test or symptom relapse within 60 days requiring retreatment) were confirmed by clinician chart review.Results: Reveles et al tool achieved the highest AUROC of 0.523 (and 0.537 among a subcohort of 1,230 patients with their first occurrence of CDI), which was not substantially better than other tools including the current IDSA/SHEA C. difficile guidelines or PCR cycle threshold (AUROC: 0.564), regardless of prior infection history.Conclusions: All tools performed poorly for predicting recurrent C. difficile infection (AUROC range: 0.488–0.564), especially among patients with a prior history of infection (AUROC range: 0.436–0.591). Future studies may benefit from considering novel biomarkers and/or higher-dimensional models that could augment or replace existing tools that underperform.
- 9 May 2024
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 45 (9) , 1064-1072
Open AccessValidation of clinical risk tools for recurrent Clostridioides difficile infection
Objective: We sought to validate available tools for predicting recurrent C. difficile infection (CDI) including recurrence risk scores (by Larrainzar-Coghen, Reveles, D’Agostino, Cobo, and Eyre et al) alongside consensus guidelines risk criteria, the leading severity score (ATLAS), and PCR cycle threshold (as marker of fecal organism burden) using electronic medical records.Design: Retrospective cohort study validating previously described tools.Setting: Tertiary care academic hospital.Patients: Hospitalized adult patients with CDI at University of Virginia Medical Center.Methods: Risk scores were calculated within ±48 hours of index CDI diagnosis using a large retrospective cohort of 1,519 inpatient infections spanning 7 years and compared using area under the receiver operating characteristic curve (AUROC) and the DeLong test. Recurrent CDI events (defined as a repeat positive test or symptom relapse within 60 days requiring retreatment) were confirmed by clinician chart review.Results: Reveles et al tool achieved the highest AUROC of 0.523 (and 0.537 among a subcohort of 1,230 patients with their first occurrence of CDI), which was not substantially better than other tools including the current IDSA/SHEA C. difficile guidelines or PCR cycle threshold (AUROC: 0.564), regardless of prior infection history.Conclusions: All tools performed poorly for predicting recurrent C. difficile infection (AUROC range: 0.488–0.564), especially among patients with a prior history of infection (AUROC range: 0.436–0.591). Future studies may benefit from considering novel biomarkers and/or higher-dimensional models that could augment or replace existing tools that underperform.