Biostatistician | Public Health Researcher | Epidemiologist
🌟 Unlocking the Power of Data in Public Health 🌟
I’m Matthew Bucala, a seasoned biostatistician and public health researcher with extensive experience in leveraging data to drive evidence-based decisions in healthcare and public policy. With a solid background in epidemiology, biostatistics, and infectious disease prevention, I specialize in transforming complex datasets into actionable insights that address pressing public health challenges. My focus is on improving healthcare quality, advancing health equity, and enhancing clinical practices through advanced data analysis.
With an MPH in Epidemiology and Biostatistics from the University of Michigan, I have honed my expertise in advanced statistical methodologies, longitudinal data analysis, and data visualization. My experience spans various domains, and I bring specialized skills in data management and programming for robust, scalable analysis, including proficiency in SQL for querying databases, Python for data manipulation and modeling, and R for statistical analysis and visualization. Some key areas include:
I believe that data has the power to tell compelling stories and uncover insights that spark meaningful change. My goal is to bridge the gap between complex data and real-world application, ensuring that research findings are not just published, but actively inform policies and practices that improve lives. Whether I’m developing statistical models, crafting data-driven reports, or presenting findings to stakeholders, my focus remains on delivering results that are impactful, actionable, and aligned with the broader goals of public health.
Over the years, I’ve had the privilege of contributing to cutting-edge projects that span academic research, clinical practice, and public health. Some of my proudest accomplishments include:
Leading Data Analysis for HIV Research: Led data management efforts at the University of Chicago, developing and executing complex SQL queries, cleaning large datasets, and automating workflows in Python and R for HIV prevention research.
Improving Infection Prevention Protocols: During my time at Ascension Health, I collaborated with clinical teams to identify gaps in infection prevention practices, leading to actionable recommendations that reduced healthcare-associated infections.
Advancing Research in Healthcare Quality: At Michigan Medicine, I led the design and analysis of quality improvement initiatives, integrating SQL for seamless data extraction and R for statistical analysis, ensuring improvements were both evidence-based and sustainable.
I am deeply motivated by the intersection of data and human stories. Behind every dataset lies a population, a community, or an individual whose experiences and outcomes can be improved through research and innovation. I strive to honor those stories by conducting rigorous analyses that shed light on disparities, inform equitable policies, and ultimately contribute to healthier communities.
I invite you to explore my website to learn more about my education, professional experience, research, and publications. If you share my passion for public health and data-driven solutions, I would love to connect and explore opportunities to collaborate.
MPH in Epidemiology/Biostatistics (2023)
Certificate in Healthcare Infection Prevention & Epidemiology
- Master’s thesis: Does it Matter Who Performs Blood Culture Collection?
BA in Anthropology (2019), Magna Cum Laude
- Honors thesis: Obesity in the Fire Service
Survival analysis is used to model time-to-event outcomes, such as survival rates or time to disease progression. Below is an example of Kaplan-Meier curves and Cox Proportional Hazards models for analyzing survival probabilities and hazard ratios.
library(survival)
library(survminer)
# Example survival dataset
data <- lung
# Kaplan-Meier estimator
km_fit <- survfit(Surv(time, status) ~ sex, data = data)
# Plot survival curves
ggsurvplot(
km_fit,
data = data,
conf.int = TRUE,
pval = TRUE,
risk.table = TRUE,
title = "Kaplan-Meier Survival Curves by Sex",
legend.labs = c("Male", "Female")
)
# Cox proportional hazards model
cox_model <- coxph(Surv(time, status) ~ age + sex, data = data)
summary(cox_model)Multistate modeling evaluates transitions between disease states, such as progression, remission, and death. This example demonstrates how to fit a multistate illness-death model.
library(mstate)
# Example dataset for illness-death model
data <- ebmt
tmat <- trans.illdeath()
# Prepare data
ms_data <- msprep(
time = c(NA, "relapse", "srv"),
status = c(NA, "relapse_status", "srv_status"),
data = ebmt, trans = tmat
)
# Fit Cox models for each transition
cox_models <- coxph(Surv(Tstart, Tstop, status) ~ age + donor, data = ms_data, id = id)
summary(cox_models)Bayesian hierarchical models incorporate random effects, allowing for precise uncertainty estimation across multiple levels. Below is an example applied to patient outcomes nested within hospitals.
library(brms)
# Simulated dataset
set.seed(123)
data <- data.frame(
hospital = rep(1:10, each = 30),
treatment = rbinom(300, 1, 0.5),
outcome = rnorm(300, mean = 10, sd = 3)
)
# Fit Bayesian mixed-effects model
bayesian_model <- brm(
outcome ~ treatment + (1 | hospital),
data = data,
family = gaussian(),
prior = c(set_prior("normal(0, 5)", class = "b")),
chains = 4, iter = 2000, control = list(adapt_delta = 0.95)
)
# Summarize and visualize results
summary(bayesian_model)
plot(bayesian_model)Mapping disease hotspots is essential for targeting public health interventions. Moran’s I quantifies spatial autocorrelation, identifying whether cases are clustered or randomly distributed.
library(spdep)
# Simulated spatial data
coords <- cbind(runif(50), runif(50))
cases <- rpois(50, lambda = 5)
data <- data.frame(x = coords[, 1], y = coords[, 2], cases)
# Create spatial neighbors
neighbors <- knearneigh(coords, k = 3)
lw <- nb2listw(knn2nb(neighbors))
# Moran's I test
moran_test <- moran.test(data$cases, lw)
moran_test
# Visualize points
plot(coords, main = "Spatial Distribution of Cases", pch = 19, cex = data$cases / 5)Explainability ensures that machine learning models are interpretable and actionable. Below is an example of using SHAP (SHapley Additive exPlanations) to interpret feature importance.
library(randomForest)
library(iml)
# Example dataset
data <- data.frame(
outcome = rbinom(200, 1, 0.5),
predictor1 = rnorm(200, mean = 5),
predictor2 = rpois(200, lambda = 3)
)
# Random Forest model
rf_model <- randomForest(outcome ~ ., data = data, ntree = 500)
# SHAP values
predictor <- Predictor$new(rf_model, data = data, y = data$outcome)
shap <- FeatureImp$new(predictor)
plot(shap)Propensity score matching adjusts for confounders in observational studies, mimicking randomized controlled trials.
library(MatchIt)
# Example dataset
data <- data.frame(
treatment = rbinom(200, 1, 0.5),
age = rnorm(200, mean = 50, sd = 10),
comorbidity_score = rpois(200, lambda = 2)
)
# Perform matching
matched_data <- matchit(treatment ~ age + comorbidity_score, method = "nearest", data = data)
summary(matched_data)
# Plot balance
plot(matched_data)Schneider J., Almirol E., English G., Chen Y., Wiger T., Bucala M., Hanson B., Matthews D., Robinson L., Pagkas-Bather A., Moody J., Moline H., Driver R., Knox J., Duncan D.
“Negative Law Interactions are Associated with Status Neutral Care in Young Black Sexual Minorities.”
(In progress)
Bouris A., Bucala M., Meyer M., Almirol E., Carter C., Vasquez L., Dawdani A., Webb J., Payne D., Johns M., Pyra M., Bouris A., Wu E., Schneider J.
“Network-based intervention to improve PrEP retention among young Black men who have sex with men: protocol for a pilot randomized controlled trial of PrEP-R (PrEP Retention).”
(Submitted)
Brewer R., Bucala M., Meyer M., Almirol E., Carter C., Vasquez L., Dawdani A., Webb J., Payne D., Johns M., Pyra M., Bouris A., Wu E., Schneider J.
“Leveraging the power of social networks to promote COVID-19 testing and vaccination among individuals who have been impacted by the criminal legal system: A mixed methods evaluation.”
(Submitted)
Bucala, M., Washington, S., Velasquez, D., Fenton, D., Bucala, M., Kalidoss, S., Faris, S.
“Career Pathways of Urology Residents: A Quantitative Cross-Sectional Analysis of Resident Career Outcomes and Influencing Factors.”
Journal of Graduate Medical Education.
(Submitted)
Wu E., Meyer M., Almirol E., Zhao X., Payne G., Bhavan K., Zaller N., Montgomery J., Hotton A., Brewer R., Johns M., Aalsma M., Knopf A., Taxman F., Hodge S., Johnson O., Carter C., Bucala M., Hazra A., Shah M., Pho M., Bouris A., Walters S., Umutoni V., Pyra M., Birkett M., Phillips G. II, Zapolski T., Webb J., Smartt J., Horton H., Durrell M., Fletcher S., Schneider J.
“A Community Network-Driven COVID-19 Testing and Vaccination Intervention for Vulnerable Populations in the Central US: Results from a RADx-UP RCT.”
Journal of Community Health.
(Submitted)
Branford G., Bucala M., Vyas J., Northway R., Hepper A., Hadeed N, Brenner M.
“Deconstructing Gender Disparities in EHR In Basket Management: A Comparative Analysis of Primary Care Providers.”
Journal of General Internal Medicine.
(Accepted, In Press)
Bucala M, Ameling J., Vyas J., Lukela J., Chrouser K.
“Unpacking Workplace Stress: Unique Factors Impacting Non-Clinical Healthcare Staff - A Narrative Review.”
Journal of Hospital Management and Healthcare Policy.
(Accepted, In Press)
Hooberman A., Ameling J., Henderson J., Manojlovich M., Bucala M., Salamey Y., Sussman J.
“Development of Multidisciplinary Team Meetings for Complex Patients: A Framework for Implementation.”
Journal of Healthcare Quality.
(Submitted)
Hadeed N., Ameling J., Henderson J., Bucala M., Salamey Y., Meddings J.
“Taming the In-Basket – How Two Simple Tools Reduced Portal Message Volume in an Academic Internal Medicine Clinic.”
Journal of General Internal Medicine.
(Submitted)
Bucala M., Hopfner D., Madigan J., Nomides N., Sharma M., Brodsky C.
“Does It Matter Who Performs Blood Culture Collection? Results of a Survey Assessing Phlebotomist, Nurse, and Resident Knowledge of Blood Culture Collection Protocols.”
Journal of Infection Prevention.
1-3.
Brodsky C., Bucala M., Rowland S., Michanowicz D.
“Burden of Natural Gas Responses on Emergency Services.”
Energy Policy.
(Submitted)
Brodsky C., Bucala M., Abdulfatah E., Siegel G.
“Extraskeletal Myxoid Chondrosarcoma: Retrospective Case Series Examining Prognostic Factors, Treatment Approaches, and Oncologic Outcomes.”
American Journal of Clinical Oncology.
46 (4), 172-177.
Woltmann E., Henderson J., Branford G., Bucala M., Lanham M., Salamey Y.
“Enhancing Team-Based In-Basket Work Using a Novel Tool Embedded in the Electronic Medical Record.”
Poster. Society of General Internal Medicine. Boston, MA. May 14–17, 2025.
Brewer R., Almirol E., Meyer M., Bucala M., Carter C., Arnold S., Vasquez L., Dawdani A., Webb J., Payne D., Pyra MM., Bouris AM., Wu E., Schneider JA.
“Social Network Diffusion of COVID-19 Prevention for Diverse Criminal Legal Involved (CLI) Communities.”
Poster. Conference on Retroviruses and Opportunistic Infections (CROI). March 9–12, 2025.
Hadeed N., Ameling J., Henderson J., Salamey Y., Bucala M., Meddings J.
“Taming the In-Basket – How Two Simple Tools Reduced Portal Message Volume in an Academic Internal Medicine Clinic.”
Poster. Society of General Internal Medicine. Boston, MA. May 15, 2024.
Hadeed N., Ameling J., Henderson J., Salamey Y., Bucala M., Meddings J.
“Taming the In-Basket – How Two Simple Tools Reduced Portal Message Volume in an Academic Internal Medicine Clinic.”
Poster. Michigan Medicine, General Medicine Symposium. Ann Arbor, MI. May 3, 2024.
Bucala M., Ameling J., Vyas J., Lukela J.
“Unpacking Workplace Stress: Unique Factors Impacting Non-Clinical Healthcare Staff - A Narrative Review.”
Poster. Michigan Medicine, General Medicine Symposium. Ann Arbor, MI. May 3, 2024.
Brodsky C.N., Bucala M., Siegel G.W.
“Extraskeletal Myxoid Chondrosarcoma: Retrospective Case Series Examining Prognostic Factors, Treatment Approaches, and Oncologic Outcomes.”
Poster. Musculoskeletal Tumor Society Annual Meeting. Clearwater Beach, FL. November 30, 2022.
Brodsky C.N., Bucala M., Siegel G.W.
“Extraskeletal Myxoid Chondrosarcoma: Retrospective Case Series Examining Prognostic Factors, Treatment Approaches, and Oncologic Outcomes.”
Poster. Michigan Orthopedic Society Scientific Meeting. Mackinac Island, MI. June 17, 2022.
Bucala M.
“Does it Matter Who Performs Blood Culture Collection? An Evaluation of Blood Culture Contamination in the Adult Intensive Care Units at Ascension St. John Hospital.”
Poster. University of Michigan School of Public Health Department of Epidemiology Poster Session. Ann Arbor, MI. November 18, 2022.
📧 mbucala@umich.edu
📞 (925) 765-1899
LinkedIn |