STATISTICAL METHOD
Survival Analysis
Statistical Consulting for Time-to-Event Data
We carry out survival analysis within your study design, fixing the time origin, the event definition and the censoring rules first, then reporting survival and hazard estimates with confidence intervals in publication-ready form.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
METHOD OVERVIEW
What Is Survival Analysis?
Survival analysis models the time until an event, such as death, relapse or device failure, when some participants have not had the event by the end of follow-up. Those participants are censored: their event time is known only to be later than their last observation, so dropping them or counting them as event-free misstates risk. The framework describes risk through the survival function, the probability of remaining event-free, and the hazard, the instantaneous event rate among those still at risk. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides survival analysis within your study design, with the relevant checks, effect estimates, uncertainty measures, interpretation and publication-ready reporting.
WHEN TO ENGAGE
When Is Survival Analysis Appropriate?
Time-to-event analysis fits any outcome defined by the time until an event, and these are the research situations in which researchers most often ask us to design, run or confirm it.
Follow-Up With Censored Patients A cohort or trial follows participants for different lengths of time, and some are still event-free when follow-up ends or contact is lost.
Defining Time Zero and the Event Diagnosis, randomization or treatment start could mark time zero, and the event could be death, relapse or a composite, so each must be fixed first.
Delayed Entry and Competing Events Patients enter observation after time zero, or another event such as death from another cause can prevent the outcome, so the standard approach needs adapting.
Reviewing a Completed Analysis A survival analysis has been run and a reviewer has queried the censoring or time scale. Independent Statistical Review checks the coding, the censoring rules and the estimates.
ANALYTICAL APPROACH
How We Apply Survival Analysis
Within the family of Clinical, Survival & Epidemiological, the question and the follow-up structure decide the model, not the other way around.
Time Origin and Time Scale We fix time zero and the event from the research question, such as randomization or diagnosis, and choose the time scale, time on study or age, before any estimate is produced.
Censoring and Follow-Up Administrative censoring at the data cutoff is rarely a concern; loss to follow-up can be, so we compare its reasons by group and test the estimates with censoring weights.
Choosing the Effect Measure We decide whether the question needs survival at a set time, a hazard ratio or restricted mean survival time, since each answers a different question about the same follow-up.
Events Limit Complexity The number of events, not the sample size, limits how many covariates a model can carry. We size the model to the events and prefer a short prespecified set to a long list.
DELIVERABLES
What You Receive
As part of our Quantitative Data Analysis service, the survival analysis is delivered with the estimates, checks and files needed to report and defend it.
Data & Variable Review Time origin, event and censoring indicators, entry times and end dates are checked for coding errors, and zero, negative or implausible follow-up times are resolved.
Method-Specific Checks Follow-up completeness, censoring patterns by group, delayed entry and competing events are examined, with proportional hazards diagnostics where a Cox model is used.
Statistical Results Survival and hazard estimates with 95% confidence intervals, median survival or restricted mean survival time, and numbers at risk, by group or covariate.
Publication-Ready Tables & Figures Survival curves with numbers at risk, event tables and model summaries, laid out to the CONSORT or STROBE items that apply or to the journal's format.
Interpretation Results are written as risk over the follow-up observed, with the censoring assumption stated and no extrapolation beyond the last observed time.
Reproducible Analysis Files R scripts, Stata do-files or SAS and SPSS syntax are provided where applicable, so curves and estimates can be regenerated when follow-up is updated.
ALTERNATIVE METHODS
When Another Method May Be More Appropriate
We choose the method by outcome and follow-up, not by habit. These approaches fit better when time is not the outcome, when rates are the target or when other event types matter.
01
Logistic Regression
A binary outcome at a fixed time point fits better when every participant is followed for the full period, so censoring does not arise and an odds ratio is the aim.
02
Poisson Regression
Event counts per unit of person-time fit when the event rate can be treated as constant over follow-up and the time pattern itself is not of interest.
03
Competing Risks Analysis
When participants can have one of several events, such as cardiac or cancer death, the analysis reports cumulative incidence for each cause instead of one curve.
04
Incidence & Prevalence Analysis
A descriptive rate of new cases per person-time, with its interval, is enough when the question is how often events occur, not how risk evolves over follow-up.
CASE EXAMPLE
How We Work in Practice
An illustration of how a time-to-event question moves from defined time zero and censoring rules to a reportable, defensible result.
ILLUSTRATIVE EXAMPLE · ONCOLOGY OUTCOMES REGISTRY · UNITED STATES
An oncology outcomes registry wanted to describe survival after first-line treatment and needed to know whether its follow-up could support the planned analysis.
The registry measured time from treatment start to death from any cause, but many patients enrolled months later, and only survivors could enroll, so counting them at risk from treatment start would overstate survival. Before any outcomes were examined, we kept time zero at treatment start, specified delayed entry so that patients contributed risk time only from enrollment, and fixed censoring at the last confirmed contact. We compared censoring across groups to judge whether it was plausibly non-informative. The prespecified summaries were the median and restricted mean survival time at a chosen horizon.
Outcome: The registry received survival estimates with 95% confidence intervals, the median where reached, restricted mean survival time at the horizon, curves with numbers at risk 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 Time-to-Event Data
Survival analyses are run in the statistical software researchers already use, with syntax or scripts provided where applicable.
RESEARCH DISCIPLINES
Time-to-Event Analysis Across Research Fields
Survival analysis appears across many research fields, each with its own event definitions, 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 a researcher who wanted a second review of the assumptions and interpretation behind a survival analysis.
“I wanted a second set of statistical eyes on our Survival Analysis, mainly because of the assumptions and interpretation behind the survival analysis rather than just obtaining an output. The review was practical and specific, and the final results were much easier to explain to the rest of the research team. The final output was concise, clearly interpreted, and ready to use in our research reporting.”
RESEARCH RESPONSIBILITY
Statistical Support Without Taking Over Your Research
We provide the statistical work for your time-to-event 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 survival 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, the clinical meaning of the findings and your conclusions remain with you, and our statistical interpretation stays within what the estimates and the follow-up 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 Survival Analysis
Short answers to what researchers ask before a time-to-event analysis, from censoring and hazards to confidentiality and the assessment process.
Why not use logistic regression for a time-to-event outcome?
Logistic regression models whether an event occurred and ignores when, so a participant followed briefly counts as a non-event and risk is understated. Survival methods use each participant’s observed time and treat censoring correctly. A binary model is reasonable only when everyone is followed for the same fixed period, and then its odds ratio answers a different question from a hazard ratio.
What is censoring, and when does it bias results?
A participant is censored when follow-up ends before the event, so we know only that the event time is later. Standard methods assume this is non-informative given the covariates: those who leave are not at systematically higher or lower risk than those who stay. The data cannot prove that, so in any censored data analysis we describe the reasons for loss and run sensitivity analyses.
Does survival analysis require proportional hazards?
No, proportional hazards is an assumption of the Cox model, not of survival analysis in general. Kaplan-Meier curves, medians and restricted mean survival time do not need it, and when it fails in a regression we consider stratification, time-varying effects or flexible parametric models instead of reporting one hazard ratio.
Can you review an analysis that has already been completed?
Yes, we can review a completed survival analysis before submission or a response to reviewers. We check the time origin, the event and censoring coding, how delayed entry and competing events were handled, the choice of summary measures and the reporting, and confirm what stands or needs to change.
How do you protect my dataset and research confidentiality?
Your registry or trial extract, follow-up files and any protocol or 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
Survival Analysis?
Share your research question, follow-up design and data. We assess whether survival 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