Preparing Patient Data for Survival Analysis: A Step-by-Step Stata Guide

At Massachusetts General Hospital, Dr. Emily Rodriguez had a big challenge. Her team was looking at how patients recover after tough surgeries. They found that old statistical methods couldn’t handle the complex time patterns¹. Survival analysis became the key to understanding patient outcomes better².

Aspect	Key Information
Definition	Survival analysis data preparation is the systematic process of structuring, cleaning, and transforming patient-level data to enable time-to-event analyses. This process involves defining event occurrences, calculating follow-up times, handling censoring, creating time-dependent variables, and ensuring data integrity. The primary purpose is to construct a dataset that accurately captures the temporal dynamics of clinical outcomes while accounting for incomplete follow-up, competing risks, and time-varying exposures—essential elements for valid survival analysis in medical research.
Mathematical Foundation	Survival analysis is built on these key mathematical functions: Survival function: \[ S(t) = P(T > t) = 1 – F(t) \] Hazard function: \[ h(t) = \lim_{\Delta t \to 0} \frac{P(t \leq T < t + \Delta t \| T \geq t)}{\Delta t} = \frac{f(t)}{S(t)} \] Cumulative hazard function: \[ H(t) = \int_{0}^{t} h(u) du = -\log S(t) \] Cox proportional hazards model: \[ h(t\|X) = h_0(t) \exp(\beta_1 X_1 + \beta_2 X_2 + … + \beta_p X_p) \] Time-dependent covariates: \[ h(t\|X(t)) = h_0(t) \exp(\beta X(t)) \] Competing risks cumulative incidence: \[ CIF_j(t) = \int_{0}^{t} S(u-) h_j(u) du \]
Assumptions	Event definition clarity: The event of interest must be unambiguously defined with clear criteria for occurrence and precise timing documentation. Non-informative censoring: Censoring mechanisms should be independent of the event process; patients lost to follow-up should not be systematically different from those who remain under observation. Time origin consistency: All subjects must have a well-defined, consistent starting point (time zero) from which follow-up time is calculated (e.g., diagnosis date, treatment initiation, study enrollment). Follow-up adequacy: Sufficient follow-up duration and minimal loss to follow-up are necessary to capture enough events for meaningful analysis; generally, events should occur in at least 10% of the study population. Data structure integrity: For time-dependent analyses, data must be correctly structured in either counting process format (start/stop times) or appropriately formatted calendar dates with accurate event indicators.
Implementation	Stata Survival Data Preparation Workflow: Data Import and Initial Inspection: `import excel "patient_data.xlsx", firstrow clear` `describe` `codebook patient_id event_date diagnosis_date death_date` Date Variable Formatting: `generate diagnosis = date(diagnosis_date, "MDY")` `format diagnosis %td` `generate death = date(death_date, "MDY")` `format death %td` Follow-up Time Calculation: `generate fu_time = (death - diagnosis)/365.25` (in years) `replace fu_time = (end_of_study - diagnosis)/365.25 if death==.` Event Indicator Creation: `generate event = 0` `replace event = 1 if death!=.` `label define event_lbl 0 "Censored" 1 "Event"` `label values event event_lbl` Handling Competing Risks: `generate cause_of_death = .` `replace cause_of_death = 1 if death_cause=="Cancer"` `replace cause_of_death = 2 if death_cause=="Cardiovascular"` `replace cause_of_death = 3 if death_cause=="Other"` Setting Up Survival Data: `stset fu_time, failure(event==1) id(patient_id) scale(365.25)` `stdes` (Describes survival data structure) Time-Dependent Covariates Setup: `stsplit time_period, at(0(1)10)` (Splits at yearly intervals) `generate treatment_status = 0` `replace treatment_status = 1 if treatment_date <= diagnosis + time_period365.25` Data Validation:* `list patient_id diagnosis death fu_time event if fu_time < 0` (Checks for negative follow-up) `list patient_id diagnosis death event if death < diagnosis` (Checks for chronological inconsistencies) Descriptive Statistics: `stsum, by(treatment_group)` `stci, by(treatment_group)` (Confidence intervals for survival times)
Interpretation	When interpreting survival data preparation outputs in Stata: stset Summary: After the `stset` command, examine the output carefully. The number of subjects, total analysis time at risk, time units, and number of failures provide an overview of your dataset's completeness. Pay attention to any "invalid observations" reported—these indicate data issues requiring resolution. Incidence Rates: The `stsum` command provides incidence rates (events per person-time). These rates offer an initial assessment of event occurrence across groups before formal modeling. Compare confidence intervals for overlap to gauge potential significant differences. Censoring Patterns: Evaluate the proportion of censored observations overall and by key subgroups. High censoring rates (>50%) in specific groups may indicate follow-up issues or selection bias. Differential censoring across comparison groups raises concerns about non-informative censoring assumption violations. Follow-up Duration: Assess median follow-up time (calculated using reverse Kaplan-Meier method with `stci, emean`) to ensure adequate observation period. Short median follow-up relative to the expected time-to-event may indicate premature analysis or insufficient maturity of data. Data Structure Verification: For time-dependent analyses using `stsplit`, verify that the resulting expanded dataset correctly represents exposure timing. Each subject should have multiple records with appropriate start/stop times and updated covariate values reflecting their status during each interval.
Common Applications	Oncology Research: Preparing cancer registry data for overall survival, progression-free survival, and disease-free survival analyses; handling multiple primary cancers; incorporating tumor characteristics and treatment response as time-dependent variables. Cardiovascular Studies: Structuring data for major adverse cardiac events (MACE) analyses; accounting for recurrent events like repeated hospitalizations; incorporating changing medication regimens and evolving risk factors. Transplantation Medicine: Preparing data for graft survival analysis; modeling competing risks of rejection, infection, and death; incorporating time-dependent immunosuppression levels and donor-specific antibody development. Infectious Disease Research: Creating datasets for analyzing time to infection clearance; handling interval-censored data from scheduled testing visits; incorporating time-varying biomarkers and antimicrobial resistance patterns. Clinical Trials: Preparing interim and final analysis datasets; implementing intention-to-treat and per-protocol approaches; handling treatment crossover; creating composite time-to-event endpoints from multiple outcome components.
Limitations & Alternatives	Complex time-dependent structures: Stata's `stsplit` approach for time-dependent covariates can become unwieldy with multiple time-varying factors. Alternatives: R's `survival` package with `tmerge` function offers more flexible handling of multiple time-dependent variables; the `flexsurv` package provides additional parametric options. Multi-state modeling limitations: While Stata can handle basic competing risks, complex illness-death or multi-state models are challenging. Alternatives: R's `mstate` package specifically designed for multi-state modeling; Python's `multistate` package for more programmatic approaches to complex transition models. Joint modeling constraints: Stata has limited built-in functionality for joint modeling of longitudinal and survival data. Alternatives: R's `JM` and `JMbayes` packages offer comprehensive joint modeling approaches; SAS PROC NLMIXED provides flexible joint modeling capabilities. Large dataset handling: Stata can struggle with very large datasets, particularly when creating expanded datasets for time-dependent analyses. Alternatives: Python with `lifelines` or `scikit-survival` packages may offer better performance for massive datasets; database-integrated approaches using SQL for initial data preparation before analysis.
Reporting Standards	When reporting survival analyses in academic publications: Clearly define the time origin (time zero) and event of interest, including precise criteria for event determination and dating. Report the number and percentage of events and censored observations overall and by key comparison groups. State the median follow-up time calculated using the reverse Kaplan-Meier method (treating the event as censoring), not simply the median observation time. For Cox regression, report hazard ratios with 95% confidence intervals and p-values, and explicitly state which covariates were included in adjusted models. Assess and report on proportional hazards assumption testing methods and results, particularly for key exposure variables. For competing risks analyses, distinguish between cause-specific hazard ratios and subdistribution hazard ratios, as they address different research questions. Present both tabular results and graphical displays (Kaplan-Meier curves or cumulative incidence curves for competing risks) with numbers at risk indicated below the x-axis at regular intervals. For time-dependent covariates, clearly describe the measurement frequency, updating mechanism, and handling of missing intermediate values. Include a STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) or CONSORT (Consolidated Standards of Reporting Trials) flow diagram detailing patient selection and exclusions.
Common Statistical Errors	Our Manuscript Statistical Review service frequently identifies these errors in survival analysis data preparation: Immortal time bias: Incorrectly classifying exposure status without accounting for the time required to receive the exposure, artificially inflating survival benefit in the exposed group. This requires proper time-dependent covariate handling rather than fixed baseline classification. Inconsistent event dating: Using different sources or criteria for determining event dates across comparison groups, introducing differential measurement error. Standardized protocols for date assignment should be implemented across all study subjects. Inappropriate handling of competing risks: Using standard Kaplan-Meier methods when competing events are present, which overestimates the probability of the event of interest. Competing risk methods (Fine-Gray model or cause-specific hazards) should be employed instead. Informative censoring ignorance: Failing to investigate or address potentially informative censoring patterns, particularly when loss to follow-up is related to prognosis. Sensitivity analyses with different censoring assumptions should be conducted. Time-scale confusion: Using time-since-entry as the time scale when age is the more relevant time dimension for the hazard function, particularly in aging-related outcomes. Consider using age as the time scale with left-truncation at study entry. Mishandling interval-censored data: Incorrectly assigning event times to scheduled visit dates rather than acknowledging that events occurred in intervals between assessments. Specialized interval-censoring methods should be applied.

Expert Services

Manuscript Statistical Review

Get expert validation of your statistical approaches and results interpretation. Our statisticians will thoroughly review your methodology, analysis, and conclusions to ensure scientific rigor.

Learn More →

Publication Support - Comprehensive assistance throughout the publication process
Manuscript Writing Services - Professional writing support for research papers
Data Analysis Services - Expert statistical analysis for your research data
Manuscript Editing Services - Polishing your manuscript for publication

Need Help With Your Statistical Analysis?

More and more, medical researchers see survival analysis as a top tool. It helps them understand the time-related aspects of health care. By using advanced stats, they can get valuable insights from Stata survival analysis data².

Time-to-event analysis is great for tracking patient progress. It captures important moments from start to finish. Now, researchers can go beyond simple methods, even with missing data¹.

Key Takeaways

Survival analysis offers comprehensive insights into time-dependent medical events
Stata provides powerful tools for sophisticated statistical modeling
Proper data preparation is crucial for accurate time-to-event analysis
Advanced statistical techniques can reveal complex patient outcome patterns
Medical research benefits from nuanced temporal data exploration

Understanding Survival Analysis in Medical Research

Survival analysis is a detailed statistical method. It helps researchers understand time-to-event data in medical studies. This tool is key for studying important health outcomes in different areas³.

At its heart, survival analysis tracks the time until a certain event happens. This could be disease progression, how well a treatment works, or when a patient dies. The Cox proportional hazards model and Kaplan-Meier estimator are two main tools used to work with these complex data sets⁴.

Definition and Importance

Survival analysis is more than just basic statistics. It deals with censored data, which is incomplete information about when an event happens. In medical studies, this is very important. It helps track patient outcomes where not all subjects experience the event of interest³.

Tracks time-dependent events in medical research
Handles incomplete or interrupted observation periods
Provides insights into patient survival probabilities

Key Concepts: Survival Time and Censoring

It's crucial to understand survival time and censoring. In medical studies, survival time is how long from diagnosis to an event. Censoring happens when we don't know the final outcome⁴.

Concept	Description
Survival Time	Period from initial observation to event occurrence
Censoring	Incomplete information about event timing

Applications in Medical Research

Survival analysis is used in many medical fields. In cancer research, it helps see how well treatments work by looking at survival rates. Clinical trials also use it to compare treatments and predict long-term outcomes⁴.

Survival analysis transforms complex medical data into meaningful insights, bridging statistical methodology and clinical understanding.

By learning the Kaplan-Meier estimator and Cox proportional hazards model, researchers can get detailed insights from censored data. This improves our medical knowledge and how we care for patients³.

Overview of Data Requirements for Survival Analysis

Survival analysis needs precise data to get accurate results. Researchers must structure their data well to capture the complex nature of time-to-event studies⁵. Knowing what's needed helps build strong analytical frameworks for medical and scientific studies.

Survival data is different from regular statistical data. The survival analysis seminar shows what researchers must think about when getting their data ready⁵.

Essential Data Types for Survival Studies

Researchers need specific variables for thorough survival analysis:

Time-to-event measurements
Censoring indicators
Subject identifiers
Covariate information

Key Variables in Survival Datasets

Good survival analysis starts with careful data preparation. Using log-rank test methods checks if the data is right⁶. Researchers should pay attention to:

Precise event timing
Accurate censoring mechanisms
Comprehensive covariate documentation

Data Quality and Preparation Strategies

Dealing with competing risks and frailty models requires careful data checking⁵. Important preparation steps include:

Identifying potential outliers
Handling missing data systematically
Ensuring data consistency

"Data preparation is the foundation of meaningful survival analysis"

By sticking to these tips, researchers can create solid datasets. These datasets support precise statistical studies in medicine and science.

Setting Up Your Dataset in Stata

Getting your dataset ready for survival analysis in Stata is a detailed process. It involves importing, cleaning, and organizing your data. This ensures your biostatistics research is accurate⁷. First, you need to transform raw patient data into a format ready for survival analysis.

Importing Data into Stata

Stata makes it easy to import data from different sources. You can use:

ASCII files
Excel spreadsheets
Other statistical package formats⁷

When you import data, check the variable types. Make sure they match Stata's needs⁸. Stata can handle big datasets, up to 32,767 variables, depending on your RAM⁸.

Cleaning and Organizing Data

Cleaning your data is essential in Stata. You'll need to:

Find and handle missing values
Change variable codes
Make new variables for survival analysis

Creating Required Variables

To do survival analysis, you must create specific variables:

Variable Type	Description
Time-to-event	Numeric variable showing survival time
Censoring indicator	Binary variable for event occurrence

Stata needs you to declare survival-time data with the stset command⁹. Correct variable creation is key for analyzing different types of censoring⁹.

Pro tip: Always check your dataset's structure before doing advanced statistical analysis.

Coding Time-to-Event Data in Stata

Survival analysis in medical research needs precise data preparation and coding. Time-to-event analysis requires careful handling of complex data for accurate results².

Researchers must grasp the key steps of coding time-to-event data in Stata. These steps turn raw medical data into survival datasets ready for analysis².

Defining the Time Variable

When preparing time-to-event data, defining the time variable is crucial. This variable shows how long it takes for an event to happen from a start point².

Select the right time measurement (days, months, years)
Choose a clear start point for time counting
Make sure all time units are the same in the dataset

Censoring and Event Indicators

Censoring is key in survival analysis. There are three main types of censoring:

Right-censoring: Events happen after the study ends
Left-censoring: The exact event time is unknown
Interval-censoring: The event is seen but exact timing is unsure²

Using Stata Commands for Recoding

Stata has strong commands for recoding survival data. The stset command is used to declare survival-time data. It takes important details like the time variable and failure indicators².

Accurate data preparation is the foundation of reliable survival analysis in medical research.

By learning these coding techniques, researchers can turn complex medical data into useful survival analysis insights¹⁰.

Selecting the Appropriate Statistical Tests

Survival analysis needs careful choice of statistical methods to understand time-to-event data well. Researchers must pick from several key methods to get reliable and useful results².

It's important to know the differences between statistical tests for accurate research. The main methods in survival analysis are the Kaplan-Meier estimator, Cox proportional hazards model, and log-rank test².

Kaplan-Meier Estimator: A Non-Parametric Approach

The Kaplan-Meier estimator is a detailed way to estimate survival probabilities. It looks at the number of people at risk and events at certain times².

Estimates survival function S(t)
Provides median and quartile survival times
Generates 95% confidence intervals

Cox Proportional Hazards Model

The Cox proportional hazards model is a strong semi-parametric method for survival data analysis. It lets researchers look at many factors at once and see how they affect survival time².

Log-Rank Test for Comparing Groups

The log-rank test is key for comparing survival curves in different groups. It shows if there are significant differences in survival patterns².

Statistical Test	Primary Purpose	Key Characteristics
Kaplan-Meier Estimator	Estimate survival probabilities	Non-parametric approach
Cox Proportional Hazards Model	Analyze multiple covariates	Semi-parametric method
Log-Rank Test	Compare survival curves	Rank-based statistical comparison

Researchers must think about their research questions and data when picking a statistical test¹¹. The right choice depends on the data type, study goals, and observation nature¹¹.

Performing Survival Analysis: Step-by-Step in Stata

Survival analysis in medical research needs precise stats to handle complex data. Stata's powerful commands help with this, tackling censored data and competing risks¹².

Stata has strong tools for survival data analysis. It supports various methods to study time-to-event outcomes¹².

Running Kaplan-Meier Analysis

The Kaplan-Meier method is a key nonparametric approach. The sts command helps generate survival estimates and show survival curves¹².

Generate survival probability estimates
Calculate median survival times
Construct confidence intervals

Conducting Cox Regression Analysis

Cox proportional hazards models let researchers look at many predictors at once. The stcox command is great for studying time-dependent covariates and frailty models¹².

Stata Command	Purpose
sts	Nonparametric survival analysis
stcox	Cox proportional hazards regression

Interpreting Stata Output

Getting Stata's output right means looking at important stats closely. Focus on hazard ratios, confidence intervals, and p-values for solid conclusions¹³.

Evaluate coefficient significance
Interpret hazard ratios
Assess model fit statistics

Learning these methods helps researchers deeply analyze survival data. They can find key insights into medical research¹².

Reporting Results from Survival Analysis

Survival analysis needs careful reporting to keep research open and precise. Researchers must present their findings clearly, using tools like Stata for biostatistics¹³.

Creating Clear and Informative Tables

Creating strong statistical tables is key. Important parts of survival analysis reporting include:

Showing hazard ratios with confidence intervals
Displaying statistical significance levels
Pointing out key variables that affect survival

Variable	Coefficient	Standard Error	Z-Value	P-Value
Age	-0.0221	0.0075	-2.95	0.003
Treatment	-0.2437	0.0905	-2.69	0.007

Visualizing Survival Curves

Graphs help us understand survival data better. Stata has tools for making survival curves that look great. This makes it easier to share complex data through pictures¹⁴.

Writing Up Findings for Publication

When writing about survival analysis, focus on:

Describing how you analyzed the data
Explaining what the stats mean
Putting your findings in context

It's also important to report on sample size, total time at risk, and how well the model fits. For example, our study had 610 subjects and 495 failures over 142,994 time units¹³.

Key Tips for Effective Data Analysis

Survival analysis needs precision and careful attention. Researchers often face challenges in time-to-event analysis. These challenges can affect their research outcomes¹⁵. It's important to know these pitfalls to keep scientific investigations reliable.

There are key areas where researchers often go wrong in survival analysis:

Improper handling of censored observations
Misspecifying the time scale
Violating Cox proportional hazards model assumptions
Misinterpreting statistical results

Common Challenges in Statistical Modeling

When using the Kaplan-Meier estimator, researchers must be careful with data preparation. Good data management strategies can make analysis more accurate¹⁵.

Challenge	Potential Impact	Recommended Solution
Censoring Errors	Biased survival estimates	Careful event classification
Sample Size Issues	Reduced statistical power	Conduct power analysis
Model Assumption Violations	Incorrect risk predictions	Diagnostic model checking

Key Strategies for Robust Analysis

Successful survival analysis needs careful attention to statistical techniques. Researchers should:

Rigorous data preprocessing
Comprehensive model diagnostics
Careful interpretation of statistical outputs
Continuous model refinement

By using these strategies, researchers can make their time-to-event analysis more reliable. This ensures more accurate scientific insights¹⁶.

Common Problem Troubleshooting

Survival analysis is complex and needs a smart way to tackle common problems. Our guide will show how to find and fix key issues that could harm the survival analysis techniques.

Addressing Missing Data Challenges

Missing data can really mess up survival analysis. It's important for researchers to have strong plans for dealing with censored data and incomplete sets. Stata data management has many tools to help with these problems:

Multiple imputation methods
Sensitivity analysis approaches
Careful examination of missing data patterns

Diagnosing Incorrect Censoring

Getting event and censoring indicators right is key for good survival analysis. A study with 3,161 participants showed how important it is to code data correctly¹⁷. Researchers should double-check their censoring indicators, focusing on:

Event timing accuracy
Proper classification of censored observations
Consistent documentation of follow-up periods

Solutions for Model Fit Problems

Fixing model fit issues needs a careful plan. The Harrell's Concordance index helps check how well models predict, scoring between 0.60 and 0.71 in survival analysis¹⁷. Important steps include:

Checking proportional hazards assumptions
Evaluating model diagnostics
Exploring alternative modeling techniques

Advanced statistical techniques can help researchers overcome complex data challenges in survival analysis.

By learning these troubleshooting methods, researchers can make sure their survival analysis results are valid and reliable. This helps bring more solid insights to medical research.

Resources for Further Learning

Stata survival analysis is complex. To get better at medical research and biostatistics, you need good learning resources. There are many educational materials to help you understand advanced statistical techniques survival analysis methodologies.

Recommended Books and Academic Resources

For researchers, there are key books and resources. They give deep insights into survival analysis techniques:

Survival Analysis: A Self-Learning Text by David G. Kleinbaum - A detailed guide on essential statistical methods¹⁸
Statistical journals focused on medical research and advanced biostatistics
Peer-reviewed publications on complex survival analysis methods

Online Stata Resources and Tutorials

There are many online platforms to improve your Stata survival analysis skills:

Official Stata documentation with detailed statistical analysis guides
Interactive online tutorials on data preparation techniques¹⁹
User-generated packages and community forums

Workshops and Specialized Courses

There are professional development opportunities for learning:

Course Type	Duration	Key Features
Online Survival Analysis Course	4 weeks	100% online, expert instruction¹⁸
Stata Intensive Training	7 weeks	Covers advanced statistical techniques²⁰

Keeping up with Stata survival analysis in medical research is key. By using these diverse resources, researchers can stay updated with new statistical methods. This helps improve their analytical skills.

Conclusion: Mastering Survival Analysis with Stata

Survival analysis is key in medical research. It helps researchers understand time-to-event data deeply. Our guide has shown how to prepare Stata survival analysis data well²¹.

By learning about data management and statistical modeling, researchers can gain valuable insights into healthcare²².

Mastering time-to-event analysis takes ongoing learning and careful attention. Researchers need to stay up-to-date with new methods and technology. Our guide shows that success in survival analysis depends on good data preparation, solid statistical methods, and understanding the research questions technical survival analysis approaches.

As medical research grows, knowing Stata survival analysis will become more important. Researchers who learn these advanced techniques will be able to make significant contributions to healthcare²¹. The secret is to stay curious, manage data well, and analyze with both skill and creativity²².

FAQ

What is survival analysis, and why is it important in medical research?

Survival analysis is a statistical method used in medical research. It helps understand how long it takes for a specific event to happen. This could be when a patient survives, a disease comes back, or a treatment works.This method is key because it deals with right-censored data. This is common in studies where not all participants experience the event during the study.

What are the key variables needed for survival analysis in Stata?

For survival analysis in Stata, you need a few key variables. These include:- Time-to-event variable: This measures how long it takes for the event to happen.- Censoring indicator: A binary variable (0/1) that shows if the event occurred.- Subject identifier: A unique ID for each participant.- Covariates: These are extra variables that might affect the outcome, like treatment group or age.

How do I handle missing data in survival analysis?

Dealing with missing data in survival analysis is important. You should:- Figure out the pattern of missingness.- Use imputation techniques like multiple imputation.- Do sensitivity analyses to see how missing data affects results.- Use Stata's commands for handling missing observations in survival models.

What is the difference between the Kaplan-Meier estimator and Cox proportional hazards model?

The Kaplan-Meier estimator is a non-parametric method. It:- Estimates survival probability over time.- Provides a simple way to see survival curves.- Doesn't account for covariates.The Cox proportional hazards model is a semi-parametric approach. It:- Allows for multiple covariates.- Estimates the impact of variables on the hazard rate.- Gives hazard ratios and confidence intervals.

How do I know if my data meets the assumptions for survival analysis?

To check if your data meets survival analysis assumptions:- Test the proportional hazards assumption with Schoenfeld residuals.- Check if continuous covariates are linear.- Look at the distribution of survival times.- Examine potential interactions between variables.- Use diagnostic plots and tests in Stata to check model assumptions.

What is censoring, and why is it important in survival analysis?

Censoring happens when we don't know the full outcome of a subject. In medical research, right-censoring is common. This is when:- The event hasn't happened by the study's end.- A subject leaves the study before the event.- Properly handling censored data ensures accurate survival probability estimates.

How can I visualize survival analysis results?

Stata offers many ways to visualize survival analysis results. You can:- Use Kaplan-Meier survival curves to show survival probability over time.- Create cumulative hazard plots to show cumulative risk.- Make forest plots to display hazard ratios for different variables.- Customize graphs with color, styling, and annotations.- Get ready-to-publish graphics for presentations and papers.

Short Note | Preparing Patient Data for Survival Analysis: A Step-by-Step Stata Guide