STATISTICAL METHOD
Competing Risks Analysis
Statistical Consulting for Competing Events
We carry out competing risks analysis within your study design, defining each event explicitly, estimating cumulative incidence by cause and choosing between cause-specific and Fine-Gray models by the question, in publication-ready form.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
METHOD OVERVIEW
What Is a Competing Risks Analysis?
A competing risks analysis estimates how often an event of interest occurs when other events, such as death from another cause, can prevent it. Treating the competing event as censoring and using one minus Kaplan-Meier overstates cumulative incidence, because participants who had it can no longer have the event of interest. The cumulative incidence function gives the probability of each event by a given time. Regression uses cause-specific hazards, the event rate among those still event-free, or the Fine-Gray subdistribution hazard for cumulative incidence. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides competing risks analysis within your study design, with the relevant checks, effect estimates, uncertainty measures, interpretation and publication-ready reporting.
WHEN TO ENGAGE
When Is Competing Risks Analysis Appropriate?
Competing risk survival analysis fits any time-to-event outcome that other events can preempt, and these are the research situations in which researchers most often ask us to apply or confirm it.
Mortality Before the Outcome In older or high-risk cohorts, death from other causes prevents outcomes such as fracture, dementia or graft failure, and shapes the incidence that is reported.
Several Causes of Failure Follow-up ends with one of several causes, such as relapse or treatment-related death, and the incidence of each must be reported beside the others.
Prediction or Mechanism Question The aim is either to predict absolute risk for a patient or to understand a treatment's effect on one event, and the two call for different models.
Competing Events Were Censored A paper censored competing events and a reviewer queried the incidence estimates. Independent Statistical Review rechecks the event definitions and the models.
ANALYTICAL APPROACH
How We Apply Competing Risks Analysis
Within the family of Clinical, Survival & Epidemiological, the estimand is chosen first, and the model follows from the question being asked.
Explicit Event Definitions We define the event of interest, every competing event and the censoring rule before modeling, and confirm that each participant has one first event or a censoring time.
Cumulative Incidence by Cause We estimate the cumulative incidence function for each cause with confidence intervals, so that the causes and event-free survival together account for every participant at each time.
Cause-Specific or Fine-Gray Cause-specific hazards suit questions about a treatment's effect on one event, and Fine-Gray suits absolute risk. We choose by the estimand, not by software, and often report both.
Interpretation and Limits A subdistribution hazard ratio is not a rate among those at risk, and a covariate can act differently on the hazard and on cumulative incidence, so we name what each estimate describes.
DELIVERABLES
What You Receive
As part of our Quantitative Data Analysis service, the competing risks analysis is delivered with the estimates, checks and files needed to report and defend it.
Data & Variable Review Event dates and cause codes are checked against the source records, ties between two event types on one date are resolved, and censoring dates are confirmed.
Method-Specific Checks Event counts by cause, the amount of censoring and covariate effects on each cause-specific hazard are examined, with the estimators compared where the difference is informative.
Statistical Results Cumulative incidence by cause with 95% confidence intervals, Gray's test across groups, and cause-specific and subdistribution hazard ratios with 95% confidence intervals.
Publication-Ready Tables & Figures Cumulative incidence plots by cause and event tables, prepared in line with STROBE or CONSORT reporting or the journal's requirements, ready to insert.
Interpretation Results are written by cause, with the competing-event definition explicit and the estimand named, so that cumulative incidence and hazards are not mistaken for each other.
Reproducible Analysis Files R, Stata or SAS scripts for the cumulative incidence, cause-specific and Fine-Gray models are provided where applicable, so each figure can be regenerated.
ALTERNATIVE METHODS
When Another Method May Be More Appropriate
Competing risks matter when other events can truly occur first. These methods fit better when they cannot, when the event type itself is the outcome or when only a descriptive rate is needed.
01
Kaplan-Meier Analysis
When competing events are rare, or the question is the event rate with competing events removed, one minus Kaplan-Meier is adequate and simpler to explain.
02
Cox Regression
When only one event can end follow-up and the aim is an adjusted hazard ratio, a standard Cox model suffices and carries no competing-event complexity.
03
Multinomial Logistic Regression
When everyone is followed to a fixed horizon, the event type at that point is a nominal outcome, and the odds of each type against none are modeled directly.
04
Incidence & Prevalence Analysis
When the aim is the burden of each cause in a population, cause-specific rates per 1,000 person-years give a direct summary that needs no regression model.
CASE EXAMPLE
How We Work in Practice
An illustration of how a competing risks question moves from explicit event definitions to estimates that are reported by cause.
ILLUSTRATIVE EXAMPLE · TRANSPLANT REGISTRY · UNITED STATES
A transplant registry studied time to acute rejection after transplantation, where death before rejection prevented the event for a share of recipients.
An earlier registry report had censored recipients at death and used one minus Kaplan-Meier, which would overstate the incidence of rejection. Before any results were examined, we defined rejection as the event of interest, death and graft loss without rejection as competing events, and the last follow-up as censoring. Because the registry wanted both the absolute risk of rejection by donor type and an understanding of whether donor type changed the rejection rate, we prespecified two models: Fine-Gray for cumulative incidence and a cause-specific Cox model for the rejection hazard, each tied to its own question.
Outcome: The registry received cumulative incidence curves by cause, subdistribution and cause-specific hazard ratios, each with 95% confidence intervals, a methods paragraph and the scripts.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Statistical Software for Competing Events
Fine-Gray and cause-specific models are run in R, Stata and SAS, with Python used for cumulative incidence estimation, and scripts are provided where applicable.
RESEARCH DISCIPLINES
Competing Events Across Research Fields
Competing risks arise in many research fields, each with its own competing events, follow-up designs and reporting conventions.
A Defined Process for Your Analysis
We review your dataset, research question and design before any work begins, so the scope, deliverables, timeline and fixed project fee are clear from the outset.
01
Share Your Research Materials
Send your dataset, research question, study design and any existing analysis plan or output. We will let you know if anything else is needed to assess the project.
02
Receive Your Assessment
Your statistically trained Research Coordinator reviews the project and confirms the required analysis, deliverables, timeline and fixed project price before work begins.
03
Analysis & Delivery
Once you confirm the scope, an in-house statistician completes the agreed analysis and delivers the outputs, interpretation and reporting files.
RESEARCHER VOICE
What Researchers Say
Feedback from an epidemiologist whose competing events made the standard survival estimates misleading.
“We knew Competing Risks Analysis was probably the right direction, but competing events that made standard survival estimates misleading was where we were stuck. OnlineSPSS reviewed the data and model rather than just rerunning the software output. The final reporting was much easier to use in the manuscript and in responding to reviewer comments.”
RESEARCH RESPONSIBILITY
Statistical Support Without Taking Over Your Research
We provide the statistical work for your competing risks analysis, and the study, its meaning and its authorship stay entirely with you.
Agreed Statistical Support We provide the agreed methodological, analytical and statistical reporting support for your competing events analysis, within the scope confirmed before work begins.
No Intellectual Contribution OnlineSPSS claims no intellectual contribution to your research. Our work is statistical support, and authorship remains yours.
Your Research Decisions The research question, which events compete and your conclusions remain with you, and our statistical interpretation stays within what the estimates and the design support.
Confidential Handling Research data and project details are handled confidentially and are not published. An NDA is available where required, and can be arranged before sensitive materials are shared.
FREQUENTLY ASKED QUESTIONS
Questions About Competing Risks Analysis
Short answers to what researchers ask before analyzing competing events, from Fine-Gray models to confidentiality and the assessment process.
Why does one minus Kaplan-Meier overstate incidence?
Treating the competing event as censoring assumes those participants stay at risk of the event of interest, as if the competing event had not happened. They cannot, so the estimate describes a hypothetical world without it. The cumulative incidence function is the real-world probability, and the incidences of all causes sum, with event-free survival, to one.
Should I use cause-specific hazards or Fine-Gray regression?
Each answers a different question. A cause-specific hazard ratio describes how a covariate changes the event rate among those still event-free, which suits etiology. Fine-Gray regression models cumulative incidence, which suits prognosis and absolute risk. The two can disagree: a covariate that raises the rate of the competing event lowers the incidence of the event of interest even when its own rate is unchanged, so reporting both often explains the result.
When do competing events matter enough to change the analysis?
They matter when they are common, as in older cohorts where death from other causes is frequent, or when the aim is absolute risk. If competing events are rare, one minus Kaplan-Meier changes little, so Kaplan-Meier or a Cox model may be enough. We check event counts by cause before recommending a model.
Can you review an analysis that has already been completed?
Yes, we can review a completed competing risks analysis before submission or a response to reviewers. We check the event definitions, whether competing events were censored, which hazard or incidence each estimate describes and the reporting by cause, and confirm what stands or needs to change.
How do you protect my dataset and research confidentiality?
Your registry or cohort files, analysis scripts and any protocol or statistical analysis plan are handled confidentially, and project information is not published. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Need Support With a
Competing Risks Analysis?
Share your research question, event definitions and data. We assess whether competing risks analysis fits your outcome and confirm the approach before any work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014