Round 16: response to reviewers

Protocol version reviewed: 0.16.0. Revised version: 0.17.0.

Verdicts: GPT-6 Astra MINOR REVISION; Fable 5.1 ACCEPT; Grok 4.7 ACCEPT. Every comment is answered below and recorded in ../comments.csv.

Correction to the round 15 response

The response to C15-1 justified the exact count likelihoods with harms outcomes (ketoacidosis, retinopathy complications, acute kidney injury). Those outcomes have a competing event, and the new likelihoods apply only to hazard based outcomes without one (at present O20) and to fixed time binomial outcomes. The competing risk routes keep their zero count restriction. GPT-6 Astra identified the error (C16-0).

Summary of main changes

• Reference arms for outcomes without a competing event: usable arm definition; one, two or more, or no usable arms; Poisson before binomial with the known status rule for f_a; scenario risks and their uncertainty from each arm own likelihood; standard error from transformed confidence limits.
• Competing risks: boundary matching after the horizon; one arm rule restated in absolute effects.
• PR02 rationale distinguishes the per year baseline from the cumulative hazard at the horizon.

GPT-6 Astra (gpt-6-astra)

C16-0 (note; Response record). The round 15 response to C15-1 cited ketoacidosis, retinopathy and acute kidney injury, but the new count likelihoods apply only to hazard based outcomes without a competing event (at present O20) and to fixed time binomial outcomes.

Response (corrected). Agreed; that justification was wrong and is corrected here. The alternative retains zero event reference arms for O20 and for fixed time binomial outcomes. It does not change the zero count restriction of the competing risk integration routes, which still applies to the harms outcomes named. Extending count likelihoods to those routes (fable-5.1, round 16) is listed for the pre-analysis amendment together with the Aalen-Johansen horizon matter of F12-note.

C16-1 (minor; Absolute effects). Reporting exclusions bypassed the one arm safeguard; contributing set for scenarios undefined; binomial input needed the known status rule.

Response (revised). Agreed; a usable reference arm covers the horizon and supplies a qualifying input, arms without one are listed and excluded from the reference risk and the scenarios, two or more usable arms fit the random effects model, one uses its likelihood with PR02 and no heterogeneity, none triggers the relative only rule, and one usable arm gives both scenarios. The binomial input uses the longest follow-up at which status is known for every randomized participant, reaching the horizon, with the 95 percent rule only in its sensitivity analysis.

C16-2 (minor; Priors). PR02 rationale did not distinguish the annual hazard from the cumulative hazard at the horizon.

Response (revised). Agreed; the rationale now states both parameterizations and the logit scale of fixed time reference arm models.

Grok 4.7 (grok-4.7)

G16-1 (minor; Absolute effects). The one arm sentence conflicted with the competing risk one arm draw.

Response (revised). Agreed; the one arm rule for outcomes without a competing event uses the arm likelihood with PR02; for outcomes with competing events the text now restates the rule of the competing risks section (tau_k not estimated, draws from N(y_a, V_a)).

G16-2 (minor; Absolute effects). An arm with person time and a follow-up proportion had two count likelihoods.

Response (revised). Agreed; Poisson when the count is reported with valid person time over one period, otherwise binomial, and an arm that reports both enters only the Poisson likelihood.

G16-3 (minor; Priors). PR02 rationale describes a per year baseline for every complementary log log use.

Response (revised). Agreed; see C16-2.

Fable 5.1 (fable-5.1)

F16-1 (minor; Absolute effects). Precedence between count likelihoods and definition of f_a.

Response (revised). Agreed; see G16-2 and C16-1 (known status rule for f_a).

F16-2 (minor; Absolute effects). Scenario risks for arms that enter by counts.

Response (revised). Agreed; your text adopted: each usable arm risk is 1 - exp(-exp(theta_a)) from its own likelihood under PR02 alone, with uncertainty from independent draws and percentiles recomputed per draw.

F16-3 (minor; Absolute effects). Scale of the delta method from a reported confidence interval.

Response (revised). Agreed; standard error from both limits transformed by log{-log(1 - .)}, half width over 1.96, or the Greenwood variance by the delta method.

F16-4 (minor; Competing risks). Boundary matching when a last interval passes the horizon.

Response (revised). Agreed; boundaries at or before the horizon must be identical, a boundary after the horizon counting as the horizon.

F16-note (note; Competing risks; priors). Reason for keeping the working normal likelihood in competing risk routes is thin; PR02 rationale clause.

Response (revised; part deferred to amendment). The reason now states only the actual choice (one multivariate normal likelihood for count and curve arms); count likelihoods for those routes are deferred to the pre-analysis amendment (see C16-0). The PR02 rationale is revised (C16-2).