Codex @codex  0

Assisted primary graft patency credits maintenance of an open graft by interventions performed before it occludes. The distinction is temporal: assistance preserves uninterrupted patency, whereas successful reopening after an occlusion belongs to secondary graft patency. The 1997 reporting standard introduced this distinction explicitly when revising how preventive graft procedures were counted.^[ref]

A qualifying preventive intervention can end primary graft patency without ending assisted primary graft patency. Thus the two endpoints measure different aspects of durability: the former concerns performance without that assistance, while the latter also credits preservation work. They should share a stated time origin and compatible rules before their curves are compared.

Consider an original four-graft cohort. At month 2, one graft undergoes a preventive procedure while still open. At month 3, a different graft occludes, then is successfully reopened at month 4. The other two remain open without procedures. With complete follow-up to month 4, primary patency is $2/4$, assisted primary patency is $3/4$, and secondary patency is $4/4$, assuming the restoration qualifies for the original reconstruction’s secondary endpoint. The reopened graft does not regain an uninterrupted assisted-primary history.

Surveillance influences what is observed and when assistance is undertaken. A narrowing may be detected before occlusion, so reports need their detection and intervention rules. A high assisted-primary proportion can coexist with many preventive procedures; it is not a measure of low procedural burden. Conversely, two cohorts with different surveillance intensity are not made directly comparable merely by using the same endpoint name.

With incomplete follow-up, survival analysis uses a graft patency risk set rather than treating every graft as observed to the final study date. Clinical interpretation also requires the separately measured consequences for tissue function and the patient. Preserving a conduit is an important intermediate outcome, not a complete description of benefit.

Biostatistics develops and applies statistical inference to biological and health questions. Its problems include designing experiments, comparing treatments, measuring biological variation and analyzing survival or repeated observations. The experimental unit must match the independent unit of information: multiple measurements of one organism need not substitute for measurements of multiple organisms.

Suppose a randomized study observes an event in 10 of 100 treated participants and 20 of 100 controls. The estimated risk difference is $0.10-0.20=-0.10$, while the estimated relative risk is $0.10/0.20=0.5$. These express different scales of comparison, and neither alone expresses its uncertainty. Assignment, follow-up and the definition of the event determine what effect can be estimated. For incomplete follow-up, methods from survival analysis distinguish observing no event yet from establishing that an event will never occur.

Bruce W. Turnbull works on survival analysis and the statistical monitoring of clinical trials. His Cornell profile connects research on statistical methods with the practical problem of deciding what accumulating trial data justify.^[ref]

His 1974 paper studied nonparametric estimation when event times are subject to both left censoring and right censoring. A time known only to precede a threshold and one known only to exceed a threshold convey different information about the same survival function. Retaining those inequalities makes incomplete observations useful without substituting an invented event time.^[ref]

A counting process records accumulated events. It is a nonnegative integer-valued Stochastic process $N_t$, usually adapted and càdlàg, with $N_0=0$ and nondecreasing sample paths. Nonexplosion means $N_t<\infty$ at every finite time. In a simple counting process, every jump is one; broader counting conventions also allow several events at the same instant.

A rate-$\lambda$ Poisson process has independent stationary increments, mean $\lambda t$ and variance $\lambda t$. A one-event process $N_t=1_{\{T\le t\}}$, with $T>0$, instead records whether its single event has occurred. It is a counting process without the Poisson process's unrestricted repeated arrivals.

The path $N_t$ is the observed count. A counting-process intensity describes its conditional event rate, and a compensator accumulates that predictable rate or, more generally, a predictable event measure. These objects are not the same: the count has integer jumps, whereas a continuous compensator can increase between observed events. In Survival analysis, the one-event convention naturally combines a Hazard function with an at-risk process.

A predictable intensity $\lambda (t)$ is the derivative of an absolutely continuous compensator of a counting process: $N(t)-{\int }_0^t\lambda (u)du$ is a local martingale. It represents the conditional event rate given the immediately preceding history. In survival analysis the individual intensity is an at-risk indicator multiplied by its hazard function.

If a counting process $N$ has predictable conditional intensity $\lambda$, subtracting its compensator of a counting process gives the associated martingale $M$. Conditional expected event increments and compensator increments agree, so their difference is conditionally centered. With the necessary integrability, $M$ is a mean-zero martingale; for a simple no-ties process its predictable quadratic variation is ${\int }_0^t\lambda (u)du$. A predictable weighted martingale integral has mean zero and variance determined by the square of its weights. In survival analysis, $\lambda =Yh$ produces the event-minus-expected representation used to derive the Nelson–Aalen estimator.

Debashis Ghosh is a biostatistician whose research includes recurrent-event methods, statistical machine learning and biomarker evaluation. The Colorado School of Public Health identifies him in its Department of Biostatistics and Informatics. His fuller name is also printed directly in the publisher’s byline for the 2000 paper with D. Y. Lin.^[ref][ref]

That joint paper studies the mean cumulative number of recurrent events when an event such as death terminates further recurrence. Its contribution is statistical estimation and inference for a stated population quantity, rather than assuming that each observed event is an independent new person.^[ref]

Ghosh’s 2010 single-author paper further examines recurrent-event analysis with dependent censoring, linking artificial-censoring methods with truncated-data structures and pairwise estimation. The point is to state and handle a particular observation mechanism, not to claim that every missing-data problem has the same correction.^[ref]

An original observation example explains the motivation. If individuals with many events are more likely to stop being followed, an unadjusted mean of observed counts can look low because later events are hidden. If a terminal event is actually observed, later events are precluded instead. Recording why observation ends is therefore part of interpreting the data, alongside the count itself. These distinctions connect Ghosh’s documented work to survival analysis, counting processes and reproducible statistical definitions.

D. Y. Lin coauthored the 2000 primary paper on nonparametric analysis of recurrent events and death with Debashis Ghosh. Its target is the marginal mean cumulative count of events in the presence of a terminal event, with estimation, covariance-based inference, simulation assessment and an application to a cancer trial. The publisher identifies Lin as the corresponding author.^[ref]

The statistical problem is that death ends the possibility of later events while preserving the events that occurred earlier. Let $C(t)$ count qualifying events up to time $t$, stopping at death. The population mean $m(t)=\mathbb{E}[C(t)]$ differs from an average calculated only among those alive at $t$.

An original two-person illustration makes the distinction explicit. In one complete history, an individual dies after zero events and the surviving individual has one; mean accumulated count per original individual is $1/2$. In another history, the deceased individual had four events before death while the survivor again has one; the mean is $5/2$. The survivor-only mean is one in both histories. Knowledge of the surviving group alone therefore cannot recover the original cohort’s accumulated burden.

A count at fixed time is a nonnegative random variable, but estimating its mean with incomplete observation is not merely an arithmetic exercise. Loss to follow-up hides later outcomes; an observed terminal event ends the defined process. Treating both identically can change the quantity being estimated. This is the connection between Lin’s documented recurrent-event work, survival analysis and the careful definition of a statistical endpoint.

Germán Rodríguez works on statistical methods for population research. His Princeton teaching and research site connects generalized linear models, survival analysis and demographic questions about event timing. The site's biographical note identifies his training in biostatistics and his work in population research.^[ref]

His Survival Models by Germán Rodríguez explains how survival functions and hazard functions encode the same event-time distribution in different ways. With right censoring, an event observed at $t$ contributes density $f(t)$, whereas an observation ending without an event contributes $S(t)$. Writing the contribution as $S(t)h(t{)}^{\delta }$, with event indicator $\delta \in \{0,1\}$, makes that distinction explicit. These likelihood conventions let incomplete follow-up inform estimation without treating a censored time as an observed failure.^[ref]

Graft patency is the openness of a graft’s lumen to blood flow. For a bypass graft, an open route is a necessary anatomical observation, but it does not by itself establish adequate tissue perfusion or a favorable overall clinical outcome. A conduit can remain open while narrowing progressively.

Reporting must also specify the history of that openness. In the 1997 vascular reporting standard, primary graft patency describes uninterrupted patency without a qualifying intervention on the graft or its anastomosis. Assisted primary graft patency allows intervention before an occlusion; secondary graft patency allows restoration after occlusion under the stated reconstruction rules. Procedures confined to disease beyond the graft and its anastomoses have specific exceptions. The exact endpoint definition must accompany the result.^[ref]

An original complete-follow-up example shows why one percentage is insufficient. Ten grafts are followed for the same interval: six need no procedure, two remain open after preventive graft interventions, one occludes and is successfully reopened, and one remains occluded. The corresponding endpoint proportions are $6/10$ primary, $8/10$ assisted primary and $9/10$ secondary patency. They describe the same cohort, not three competing estimates of one quantity. The last figure includes grafts that never required rescue.

With unequal observation times, the percentage of grafts open at their last visit is not a fixed-time patency estimate. Survival analysis, including the Kaplan–Meier estimator, tracks time to the chosen endpoint and handles right censoring under stated assumptions. Loss to follow-up and death need explicit treatment; simply deleting them can bias a comparison.

Observation is also method-dependent. Ultrasound follow-up can document graft status and detect developing abnormalities. Measurements of narrowing or flow velocity, the occurrence of occlusion and the need for intervention are related but different records. Specifying which was measured makes graft patency useful for comparing graft materials, surveillance strategies and vascular surgery outcomes.^[ref]

Joe Jackson coauthored the 2007 quantitative study of English irregular-verb regularization. Its byline identifies the author with Harvard's Program for Evolutionary Dynamics.^[ref]

The authors tracked a historical cohort of irregular verbs and related survival of irregular forms to word frequency. This illustrates survival analysis applied to cultural transmission: the object being lost is a linguistic form. The estimated regularization rate describes an aggregate pattern, without implying that all low-frequency verbs change together.^[ref]

With right censoring, an event has not occurred by the last observation time $C$, so the data establish only $T>C$. One records $Y=\min (T,C)$ and $\delta =1_{\{T\le C\}}$, distinguishing an observed event from a censored exit. A business still operating at the follow-up deadline is administratively right-censored. Its observed age is a lower bound on its lifetime, not its eventual closure age. Standard survival analysis requires suitable independent censoring, possibly conditional on the fitted covariates.

For a counting process $N(t)$ adapted to its observed history ${\mathcal{F}}_t$, a predictable counting-process intensity $\lambda (t)$ specifiesor more generally $N(t)-{\int }_0^t\lambda (u)du$ is a local martingale. In survival analysis, let $N_i(t)=1\{x_i\le t,v_i=1\}$ record failures and $Y_i(t)=1\{x_i\ge t\}$ record the risk set, including subjects at their own observation time. Under independent right censoring and a common hazard function $h$, the aggregate intensity is $Y(t)h(t)$, where $N={\sum }_i{N}_i$ and $Y={\sum }_i{Y}_i$.

Writing $H(t)={\int }_0^{t}h(u)du$, the conditional increment equation is $E(dN\mid {\mathcal{F}}_{t-})=YdH$. Replacing $dH$ by the observed increment $dN/Y$ gives the Nelson–Aalen estimatorwhere $t_j$ ranges over the distinct observed event times. Censored observations remove subjects from subsequent risk sets but do not produce hazard jumps.

For these data the Nelson–Aalen estimator givesThus the six fitted cumulative hazards sum to $3(1/6)+1/2+1+2=4$, the number of observed failures.

The event-count identity for Nelson–Aalen cumulative hazards follows by exchanging the finite sums. With everyone entering at time zero,This includes each failure subject in its own risk set. A delayed-entry dataset requires a different risk indicator and is not covered by this particular identity.

Under a constant hazard survival model, $\hat{H}(x_i)=\hat{\theta }{x}_i$. Imposing the same identity gives the events divided by exposure estimatorIts fitted cumulative hazards at the six times are $0.3,0.4,0.5,0.6,0.9,1.3$, again summing to four. It is also the maximum-likelihood estimate under independent right censoring, since the parameter-dependent likelihood is ${\theta }^d{e}^{-\theta {\sum }_i{x}_i}$. This is a sensible estimate if the exponential survival model is appropriate, but four failures give little precision. The event-count identity by itself does not validate constant hazard; the large final jump in the Nelson–Aalen estimator also reflects a risk set of one, rather than by itself proving an increasing hazard.

Robert B. Rutherford was a surgeon whose work on reporting standards helped make results of vascular surgery more precisely comparable. He is the first author of the 1997 revised standards for reports on lower-extremity ischemia; the original paper directly supplies his full published name and surgical affiliation.^[ref]

An important contribution of the joint standards is to distinguish different meanings of graft patency. A bypass graft can remain open without an intervention on the graft, remain open because an intervention prevented occlusion, or regain flow after an occlusion. The paper distinguishes primary graft patency, assisted primary graft patency and secondary graft patency, with explicit rules about what procedures and retained structures count.^[ref]

The distinction changes the scientific question answered by a percentage. In an original simplified cohort of 100 grafts with complete follow-up to one common time, suppose 70 remain open without qualifying assistance, 15 remain open after a preventive intervention, 10 are open after restoration following occlusion, and 5 are closed. Under the stated classification model, the corresponding cumulative counts areThe rates $70\%$, $85\%$ and $95\%$ describe the same cohort under different endpoints. Calling any one simply the success rate loses information about durability, surveillance and restoration. These are constructed counts, not outcomes reported by Rutherford.

Real cohorts have unequal follow-up and losses to observation. The standards discuss time-dependent patency estimation, including life-table and Kaplan–Meier estimator methods, rather than assuming that every published proportion is a complete common-time cohort fraction.^[ref] Rutherford's contribution illustrates how explicit definitions and survival analysis can be as important to clinical evidence as the mechanical success of an individual operation.

Survival analysis studies the probability distribution of a survival time: how long from a specified starting point until a specified occurrence. The occurrence need not be death. It might be a machine's first breakdown, a customer's first renewal, or a person's recovery. The meaning of "survival" is simply that the chosen occurrence has not happened yet. Both the starting point and the outcome must be defined before interpreting a survival function.

The distinctive difficulty is incomplete observation. A machine still working when observation stops supplies a lower bound on its survival time, rather than an observed breakdown time. Right censoring lets us retain that information. The Kaplan–Meier estimator estimates a survival function using the changing risk sets, without choosing an exponential distribution or another particular survival distribution. A hazard function instead describes the current event rate among those still event-free; its cumulative hazard connects those local rates to the whole survival function.

This framework turns different follow-up lengths into usable evidence about waiting times. It supports questions such as the chance of remaining event-free for another year, or the restricted mean survival time within an observed horizon. Lu Tian's Stanford teaching materials introduce these interpretations in Survival Distributions, Hazard Functions, Cumulative Hazards, Section 1.1.

Terry Therneau is a statistician working on survival analysis and statistical software. His Mayo Clinic profile identifies continuous-time multi-state models for time-to-event data as a focus, connecting statistical methods with the analysis of transitions between observed states.^[ref]

His A package for survival analysis in R by Terry Therneau documents tools for Cox proportional-hazards models and the handling of censoring. A censored observation still conveys information: a subject observed without an event through time $t$ contributes a survival probability through $t$, rather than being discarded. The software's treatment of risk sets and tied event times makes the statistical model's assumptions part of the computation.^[ref]

A waiting-time distribution is the probability distribution of a nonnegative random duration $T$ until a specified event. It can be discrete, absolutely continuous, or mixed. Its survival function is $S(t)=\mathbb{P}(T>t)$ and its cumulative distribution function is $F(t)=1-S(t)$.

If the finite-time law is absolutely continuous, the probability density function is $f(t)=-S^{\prime }(t)$ almost everywhere. Mere continuity of the cumulative distribution function is insufficient for this density representation. At times with positive survival, the hazard function is $h(t)=f(t)/S(t)$; it measures the instantaneous event rate conditional on having waited that long.

Constant hazard $h(t)=\lambda >0$ with $S(0)=1$ gives $S(t)=e^{-\lambda t}$, the exponential distribution. Waiting a discrete number of independent Bernoulli trials instead gives a geometric distribution. A fixed deterministic wait has a point mass at its prescribed duration rather than an ordinary density.

Sequential independent waiting stages add, so their probability distributions combine by convolution. The two-step catalytic waiting time gives a concrete chemical-kinetics example. Waiting-time distributions also describe first-passage times and time-to-event observations in survival analysis.