Home / Disciplines /

Medical research

Medical researchers rely on Stata for its range of biostatisticalmethods, reproducibility, and ease of use. Whether you are conductingbasic medical research or carrying out a clinical trial, Stata providesthe tools you need to conduct your study from power and sample-sizecalculations to data management to analysis.

Features for medical researchers

General linear models
Fit one- and two-way models. Or fit models with three, four,or even more factors. Analyze data with nested factors, with fixed andrandom factors, or with repeated measures. Use ANCOVA models when you havecontinuous covariates and MANOVA models when you have multiple outcomevariables. Further explore the relationships between your outcome andpredictors by estimating effect sizes and computing least-squares and marginalmeans. Perform contrasts and pairwise comparisons. Analyze and plotinteractions.

Linear, binary, and count regressions
Fit classical ANOVA and linear regression models of the relationship between acontinuous outcome, such as weight, and the determinants of weight, such asheight, diet, and level of exercise. If your response is binary, ordinal,categorical, or count, don't worry. Stata has estimators for these types ofoutcomes too. Use logistic regression to estimate odds ratios. Estimateincidence rates using a Poisson model. Analyze matched case–control data withconditional logistic regression. A vast array of tools is available afterfitting such models. Predict outcomes and their confidence intervals. Testequality of parameters. Compute linear and nonlinear combinations ofparameters.

Power, precision, and sample size
Before you conduct your experiment, determine the sample size needed to detectmeaningful effects without wasting resources. Do you intend to compute CIs formeans or variances or perform tests for proportions or correlations? Do youplan to fit a Cox proportional hazards model or compare survivor functionsusing a log-rank test? Do you want to use a Cochran—Mantel—Haenszel test ofassociation or a Cochran—Armitage trend test? Use Stata'spower command tocompute power and sample size, create customized tables, and automaticallygraph the relationships between power, sample size, and effect size for yourplanned study. Or use theciwidthcommand to do the same but for CIs insteadof hypothesis tests by computing the required sample size for the desired CIprecision. Or usegsdesignto compute stopping boundaries and the required sample sizes for group sequentialdesigns. Instead of commands, use the interactive Control Panel to perform your analysis.

Marginal means, contrasts, and interactions
Marginal means and contrasts let you analyze the relationships between youroutcome variable and your covariates, even when that outcome is binary, count,ordinal, categorical, or survival. Compute adjusted predictions with covariatesset to interesting or representative values. Or compute marginal means foreach level of a categorical covariate. Make comparisons of the adjustedpredictions or marginal means using contrasts. If you have multilevel dataand random effects, these effects are automatically integrated out to providemarginal (that is, population-averaged) estimates. After fitting almost anymodel in Stata, analyze the effect of covariate interactions, and easily createplots to visualize those interactions.

Multilevel mixed-effects models
Whether the groupings in your data arise in a nested fashion (patients nestedin clinics and clinics nested in regions) or in a nonnested fashion (regionscrossed with occupations), you can fit a multilevel model to account for thelack of independence within these groups. Fit models for continuous, binary,count, ordinal, and survival outcomes. Estimate variances of random interceptsand random coefficients. Compute intraclass correlations. Predict randomeffects. Estimate relationships that are population averaged over the randomeffects.

Meta-analysis
Combine results of multiple studies to estimate an overall effect. Useforest plots to visualize results. Use subgroup analysis andmeta-regression to explore study heterogeneity. Use funnel plots andformal tests to explore publication bias and small-study effects. Usetrim-and-fill analysis to assess the impact of publication bias onresults. Perform cumulative and leave-one-out meta-analysis. Performunivariate, multilevel, and multivariate meta-analysis. Use themeta suite, or let the Control Panel interfaceguide you through your entire meta-analysis.

Multiple imputation
Account for missing data in your sample using multiple imputation. Choose fromunivariate and multivariate methods to impute missing values in continuous,censored, truncated, binary, ordinal, categorical, and count variables. Then, in a single step, estimate parameters using the imputed datasets, and combineresults. Fit a linear model, logit model, Poisson model, hierarchical model,survival model, or one of the many other supported models. Use themi command,or let the Control Panel interface guide you through your entire MI analysis.

Survival analysis
Analyze duration outcomes—outcomes measuring the time to an eventsuch as failure or death—using Stata's specialized tools forsurvival analysis. Account for the complications inherent in survivaldata, such as sometimes not observing the event (right-, left-, andinterval-censoring), individuals entering the study at differing times(delayed entry), and individuals who are not continuously observedthroughout the study (gaps). You can estimate and plot the probabilityof survival over time. Or model survival as a function of covariatesusing Cox, Weibull, lognormal, and other regression models. Predicthazard ratios, mean survival time, and survival probabilities. Do youhave groups of individuals in your study? Adjust for within-groupcorrelation with a random-effects or shared-frailty model. If you havemany potential covariates, uselasso cox andelasticnet cox for model selection and prediction.

Epidemiological tables
Want to analyze data from a prospective (incidence) study, cohort study,case–control study, or matched case–control study? Stata's tables forepidemiologists make it easy to summarize your data and compute statisticssuch as incidence-rate ratios, incidence-rate differences, risk ratios, riskdifferences, odds ratios, and attributable fractions. You can analyzestratified data too—compute Mantel–Haenszel combined estimates, performtests of homogeneity, and standardize estimates. If you have an ordinal ratherthan binary exposure, you can perform a test for a trend.

Additive models of relative risk
Determine how exposures interact to put subjects at a higher risk ofexperiencing an outcome of interest. For example, you might beinvestigating how exposure to cigarette smoke and asbestos interact toincrease the risk of lung cancer. With Stata'sreri command, youcan measure two–way interactions in an additive model of relativerisk, while accounting for other risk factors. Choose from varioussupported models, such as binomial generalized linear, Poisson, negativebinomial, logistic, Cox, parametric survival, andinterval–censored parametric and semiparametric survival models.Estimate the relative excess risk due to interaction (RERI),attributable proportion (AP), and synergy index (SI).

Automated reporting and customizable tables
Stata is designed for reproducible research, including the ability tocreate dynamic documents incorporating your analysis results. CreateWord or PDF files, populate Excel worksheets with results and formatthem to your liking, and mix Markdown, HTML, Stata results, and Statagraphs, all from within Stata. Create tables that compare regression results or summary statistics, use default stylesor apply your own, and export your tables to Word, PDF, HTML, LaTeX,Excel, or Markdown and include them in your reports.

Jupyter Notebook with Stata
Jupyter Notebook is widely used byresearchers and scientists to share their ideas and results for collaborationand innovation. It is an easy-to-use web application that allows you tocombine code, visualizations, mathematical formulas, narrative text, and otherrich media in a single document (a "notebook") for interactive computing anddeveloping. You can invoke Stata and Mata from Jupyter Notebook with theIPython (interactive Python) kernel. Thismeans you can combine the capabilities of both Python and Stata in a singleenvironment to make your work easily reproducible and shareable with others.

I've used a lot of stat packages over the years, but I find that I'm using Stata 95% of the time now. It's wonderful! Its speed and power are much touted, but its simplicity for beginners is perhaps one of its best features.

— Rodney Hayward
University of Michigan's Schools of Medicine & Public Health, Ann Arbor VA's Center for Clinical Management Research

Check out Stata'sfull list of features, or seewhat's new in Stata 18.

Why Stata?

Intuitive and easy to use.
Once you learn the syntax of one estimator, graphics command, or data management tool, you will effortlessly understand the rest.

Accuracy and reliability.
Stata is extensively and continually tested. Stata's tests produce approximately 6 million lines of output. Each of those lines is compared against known-to-be-accurate resultsacross editions of Stata and every operating system Stata supports toensure accuracy and reproducibility.

One package. No modules.
When you buy Stata, you obtaineverything for your statistical,graphical, and data analysis needs. You do not need to buy separate modules or import your data to specialized software.

Write your own Stata programs.
You can easily write your own Stata programs and commands. Share them with others or use them to simplify your work. Utilize Stata's do-files, ado-files, and Mata: Stata's own advanced programming language that adds direct support for matrix programming. You can also access and benefit from the thousands of existing Stata community-contributed programs.

Extensive documentation.
Stata offers 36 manuals with more than 19,000 pages of PDF documentation containing detailed examples, in-depth discussions, references to relevant literature,and methods and formulas. Stata's documentation is a great place to learn about Stata and the statistics, graphics, data management, and data science tools you are using for your research.

Top-notch technical support.
Stata's technical support is known for their prompt, accurate, detailed, and clear responses. People answering your questions have master's and PhD degrees in relevant areas of research.

Learn more

Would you like to see Stata in action?

Join us for one of our free live webinars.Ready. Set. Go Stata shows you how to quickly get started manipulating, graphing, and analyzing your data. Or, go deeper in one of ourspecial-topics webinars.

Would you like to see more?

Stata's YouTube has over 300 videos with a dedicated playlist of methodologies important to medical researchers. And they are a convenient teaching aid in the classroom.

Visit our channel

NetCourses: Online training made simple

Get started quickly at using Stata effectively, or even learn how to perform rigorous time-series, panel-data, or survival analysis, all from the comfort of you home or office.NetCourses make it easy.

For Stata users, by Stata users

Stata Press offers books with clear, step-by-step examples that make teaching easier and that enable students to learn and medical researchers to implement the latest best practices in analysis.