# Response to Reviewers: [Manuscript ID and short title]

> Point-by-point response letter for a revised journal manuscript: a cover
> note to the editor, a summary of the main changes, every editor and
> reviewer comment quoted and answered in turn, a change log that maps each
> comment to the exact place in the revised manuscript, and a final check
> before resubmission. Based on common editorial practice and the
> transparency and courtesy principles of the COPE Core Practices, cited as
> a reference only. The journal's own revision instructions always take
> precedence. Replace every `[bracketed]` value and every row marked
> *Example*.

<!--
Guidance: How to use this template
- Copy every editor and reviewer comment into section 3 before you write a
  single answer. Split long comments into numbered parts (R1.3a, R1.3b) so
  nothing is missed.
- Give each comment an ID (E1, R1.1, R2.4) and use the same ID in the
  change log (section 4), in tracked changes and in the cover note.
- Answer every comment, including praise and comments you disagree with.
  An unanswered comment is the most common reason a revision is sent back.
- Quote the new or changed text and give its location in the revised
  manuscript (section, page, line). "We have revised the paper" is not an
  answer.
- Keep the tone factual and polite. Thank reviewers once, at the top, not
  in every reply.
- Related templates: Peer Review Report (`academic/peer-review`), Research
  Paper (IMRaD) (`academic/research-paper-imrad`), Manuscript Cover Letter
  (`academic/manuscript-cover-letter`).
-->

| Field | Value |
|---|---|
| Manuscript ID | [NRPH-2026-0187] |
| Title | [Revised manuscript title] |
| Journal | [Journal name] |
| Handling editor | [Editor name] |
| Decision received | YYYY-MM-DD ([Major revision / Minor revision]) |
| Revision deadline | YYYY-MM-DD |
| Revision round | [Revision 1] |
| Corresponding author | [@author] |
| Files submitted | [Response letter, revised manuscript (clean), revised manuscript (tracked changes), revised supplement] |

## 1. Cover note to the editor

<!-- Guidance: One short letter (150 to 300 words). Thank the editor and reviewers, state that every comment has been answered, name the two or three most important changes, and flag any comment you could not fully address and why. The editor often reads only this page before deciding whether to send the paper back to reviewers. -->

Dear [Dr Editor name],

Thank you for the opportunity to revise our manuscript "[Title]"
([Manuscript ID]) and for the careful comments from you and the
reviewers. We have answered every comment below, quoting each one in
italics followed by our reply and the location of the change in the
revised manuscript.

The main changes are:

1. [*Example:* We re-ran the primary analysis with a model that accounts
   for repeated visits by the same patient, as requested by Reviewer 1
   (R1.1). The direction and size of the main estimate are unchanged.]
2. [*Example:* We added a sensitivity analysis excluding the 2025 heat
   season, when a new triage system was introduced (R2.2).]
3. [*Example:* We shortened the Discussion by about 400 words and removed
   the policy recommendation that went beyond our data (E2, R2.3).]

[*Example:* We respectfully did not add the multi-city comparison
suggested in R1.4, because data for the other cities are not available
under our ethics approval. We explain this in our reply and have added it
as a limitation.]

All page and line numbers refer to the revised manuscript with tracked
changes. [We confirm that all authors have approved the revised
manuscript.]

Yours sincerely,
[Corresponding author name, on behalf of all authors]

## 2. Summary of changes

<!-- Guidance: A table the editor can scan in one minute. Count comments by status so the editor sees the whole picture. Make sure the counts match section 3 and the change log exactly. -->

| Source | Comments | Changed as requested | Changed in part | Not changed (reason given) |
|---|---|---|---|---|
| Editor *(Example)* | 2 | 2 | 0 | 0 |
| Reviewer 1 *(Example)* | 4 | 2 | 1 | 1 |
| Reviewer 2 *(Example)* | 3 | 3 | 0 | 0 |
| **Total** *(Example)* | **9** | **7** | **1** | **1** |

Other changes not requested by reviewers:

- [*Example:* Corrected a typing error in Table 2 (the number of visits
  in the reference group was 12,480, not 12,840); no estimate changed.]
- [*Example:* Updated two references that have since been published.]

## 3. Point-by-point responses

<!-- Guidance: Use the same layout for every comment: the comment in italics (verbatim, or lightly trimmed with "[...]" where long), then "Response", then "Change" with location and the quoted new text. When you disagree, say so plainly, give the evidence, and offer what you can do instead (a limitation, a sensitivity analysis, a sentence of clarification). Never argue with the reviewer's motives. -->

### 3.1 Editor

**E1.** *[Paste the editor's comment. Example: "Please follow the
journal's reporting guideline for observational studies and upload the
completed checklist."]*

**Response.** [*Example:* Thank you. We have completed the STROBE
checklist for cohort studies and uploaded it as Supplementary File 3.
Completing it showed that the handling of missing temperature data was
not described, so we added a sentence to the Methods.]

**Change.** [*Example:* Methods, section 2.5, page 7, lines 152-155:
"Daily maximum temperature was missing for 3 of 1,096 study days; these
days were excluded from the analysis."]

**E2.** *[Example: "The Discussion is long and the final paragraph makes
policy recommendations that the data cannot support."]*

**Response.** [*Example:* We agree. We removed the final paragraph and
shortened the Discussion from about 1,900 to about 1,500 words.]

**Change.** [*Example:* Discussion, pages 15-18; deleted paragraph shown in
tracked changes on page 18.]

### 3.2 Reviewer 1

[*Example:* We thank Reviewer 1 for the detailed reading of our methods.]

**R1.1.** *[Example: "Patients may visit the emergency department more
than once. The model treats visits as independent, which underestimates
the standard errors."]*

**Response.** [*Example:* This is a fair point. We refitted the model with
standard errors clustered by patient. The main estimate for adults aged
75 and over is unchanged (rate ratio 1.12) and the 95% confidence interval widened slightly, from
1.08-1.16 to 1.07-1.17.]

**Change.** [*Example:* Methods, section 2.6, page 8, lines 170-176
(new text on clustering); Results, Table 3, page 11; Abstract, line 18.]

**R1.2.** *[Comment]*

**Response.** [Response]

**Change.** [Location and quoted new text]

**R1.3.** *[Example: "Please justify the choice of a 3-day lag window."]*

**Response.** [*Example:* We chose the 3-day window before analysis, based
on the protocol registered on [registry and date]. We kept it as the main
analysis and added 0-day and 0-to-7-day windows as a sensitivity
analysis. Only part of the request could be met: the protocol did not
specify a distributed lag model, so we report it in the supplement and
not as a main result.]

**Change.** [*Example:* Methods, page 8, lines 178-181; Supplementary
Table S4.]

**R1.4.** *[Example: "A comparison with other cities would strengthen the
generalisability."]*

**Response.** [*Example:* We agree that this would be valuable. However,
our ethics approval covers only records from the two hospitals in the
study area, and we have no lawful access to records from other cities.
We have added this as a limitation and suggest it for future work.]

**Change.** [*Example:* Discussion, Limitations, page 17, lines 361-364.]

### 3.3 Reviewer 2

**R2.1.** *[Comment]*

**Response.** [Response]

**Change.** [Location and quoted new text]

**R2.2.** *[Example: "A new triage system was introduced in 2025. Could
this explain part of the effect?"]*

**Response.** [*Example:* Thank you for raising this. We repeated the
analysis excluding the 2025 heat season. The estimate was similar (rate
ratio 1.11, 95% CI 1.05-1.17).]

**Change.** [*Example:* Results, section 3.4, page 13; Supplementary
Table S5.]

**R2.3.** *[Comment]*

**Response.** [Response]

**Change.** [Location and quoted new text]

## 4. Change log

<!-- Guidance: One row per comment ID, in the same order as section 3. This is the traceability table: an editor or reviewer can check each change without rereading the letter. Status is one of Changed, Changed in part, Not changed. Every "Not changed" row needs a reason in section 3. -->

| Comment ID | Topic | Status | Location in revised manuscript | Files affected |
|---|---|---|---|---|
| E1 *(Example)* | Reporting checklist | Changed | Methods 2.5, p7 l152-155 | Manuscript, Supplementary File 3 |
| E2 *(Example)* | Discussion length | Changed | Discussion, p15-18 | Manuscript |
| R1.1 *(Example)* | Repeated visits | Changed | Methods 2.6, p8 l170-176; Table 3; Abstract l18 | Manuscript |
| R1.2 *(Example)* | [Topic] | Changed | [Location] | [Files] |
| R1.3 *(Example)* | Lag window | Changed in part | Methods p8 l178-181; Table S4 | Manuscript, Supplement |
| R1.4 *(Example)* | Other cities | Not changed (limitation added) | Discussion p17 l361-364 | Manuscript |
| R2.1 *(Example)* | [Topic] | Changed | [Location] | [Files] |
| R2.2 *(Example)* | Triage change | Changed | Results 3.4, p13; Table S5 | Manuscript, Supplement |
| R2.3 *(Example)* | [Topic] | Changed | [Location] | [Files] |

```mermaid
flowchart LR
  A["Decision letter received"] --> B["Split comments and assign IDs"]
  B --> C["Agree a reply and owner per comment"]
  C --> D["Revise manuscript with tracked changes"]
  D --> E["Write responses and change log"]
  E --> F{"Every ID answered and located?"}
  F -- "No" --> C
  F -- "Yes" --> G["All authors approve"]
  G --> H["Resubmit"]
```

## 5. Working plan for co-authors

<!-- Guidance: Internal only: delete before submission. Assign an owner and a due date to each comment so the revision finishes before the journal deadline. Agree early on comments you plan to decline, because these need the most careful wording. -->

| Comment ID | Owner | Planned reply | New analysis needed | Due |
|---|---|---|---|---|
| R1.1 *(Example)* | [@statistician] | Refit with clustered errors | Yes | YYYY-MM-DD |
| R1.4 *(Example)* | [@lead-author] | Decline, add limitation | No | YYYY-MM-DD |
| [ID] | [@owner] | [Plan] | [Yes / No] | YYYY-MM-DD |

## 6. Writing good responses

<!-- Guidance: Read once before drafting. These habits make the difference between a second round and an acceptance. -->

- **Answer the question asked.** Start each reply with a direct answer
  (agree, partly agree, disagree) before the explanation.
- **Show, do not tell.** Quote the new sentence or give the new number.
  Reviewers should not need to search the manuscript.
- **Disagree with evidence.** Cite data, the registered protocol, the
  reporting guideline or published methods work. Offer a middle path where
  one exists, such as a sensitivity analysis or a limitation.
- **Keep numbers consistent.** When an analysis changes, update the
  Abstract, Results, tables, figures and supplement together, and say so in
  the reply.
- **Handle conflicting reviewers openly.** When two reviewers ask for
  opposite changes, say which you followed and why, and refer the editor to
  both replies.
- **Do not add new claims quietly.** Any new analysis or conclusion goes in
  the change log so reviewers can judge it.
- **Respect confidentiality.** Do not share the reviewers' comments outside
  the author group, and do not try to identify anonymous reviewers.
- **Disclose tools.** If you used AI-assisted tools to draft replies or
  revise text, follow the journal's policy and disclose the use as it
  requires; the authors remain responsible for every statement.

## 7. Pre-resubmission checklist

<!-- Guidance: Tick every line before uploading. The last three items are the ones most often forgotten. -->

- [ ] Every editor and reviewer comment has an ID and a reply in section 3
- [ ] Section 2 counts match section 3 and the change log
- [ ] Every "Change" entry gives section, page and line in the tracked file
- [ ] Every declined request gives a reason and, where possible, an
      alternative
- [ ] Abstract, tables, figures and supplement match the revised Results
- [ ] Clean and tracked-changes versions are both uploaded, as the journal
      asks
- [ ] Reporting checklists and declarations are updated if the content
      changed
- [ ] All authors have read and approved the revision and the letter
- [ ] Internal working plan (section 5) is deleted
- [ ] Line numbers were checked after the final edit

## 8. References

<!-- Guidance: List only the sources you cite in your replies, in the journal's reference style. Add any new references to the manuscript reference list too. -->

1. [*Example:* STROBE Statement, checklist for cohort studies, cited in E1.]
2. [Reference cited in a reply]

## Revision history

<!-- Guidance: Track your own drafts of the letter so co-authors know which version they are commenting on. -->

| Version | Date | Author | Change |
|---|---|---|---|
| 0.1 | YYYY-MM-DD | [@author] | Comments split and numbered |
| 0.2 | YYYY-MM-DD | [@author] | Draft replies for co-author review |
| 1.0 | YYYY-MM-DD | [@author] | Approved by all authors and submitted |

## Standard and licence note

Based on common editorial practice for revised manuscripts, with the
principles of transparency, fairness and courtesy in the COPE (Committee on
Publication Ethics) Core Practices, including peer review and complaints
and appeals, cited as a reference only (https://publicationethics.org/core-practices).
The reviewer confidentiality item reflects the COPE *Ethical Guidelines for
Peer Reviewers* (https://doi.org/10.24318/cope.2019.1.9). The AI-use item
reflects the ICMJE *Recommendations* (updated January 2026,
https://www.icmje.org/recommendations/). Journals set their own revision
rules, so follow the decision letter and author instructions, and check the
current COPE and ICMJE texts. This template is original text and
reproduces no COPE or ICMJE wording. Template licence: MIT.