REVIEW 3 major objections 4 minor 1 cited by
From Patient Burdens to User Agency: Designing for Real-Time Protection Support in Online Health Consultations
T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read The paper claims SafeShare, a real-time redaction tool built on a localized language model, lets patients remove personally identifiable details from online medical consultations before anything is sent, with a reported 89.64% detection…
desk verdict The qualitative findings and the SafeShare concept are worth your attention, but the headline accuracy number is not trustworthy because the gold labels were produced by an advanced LLM. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing machinery is the localized LLM-based PII detection module inside SafeShare. It performs selective anonymization: it identifies personally identifiable information in free-text consultation messages and replaces or removes those spans in real time, leaving the medical content intact. A 4-billion-parameter model is reported to reach 89.64% accuracy on the IMCS21 evaluation set.
What would settle it
Have multiple physicians independently annotate a random sample of IMCS21 consultations and compare their labels to the PII detection module's output; if human agreement is substantially lower than the reported 89.64%, the accuracy claim is not supported.
Extended reading notes
Core claim
SafeShare is an interaction technique that lets a user see and control what is shared: a localized LLM detects private fields such as names, phone numbers, hospital identifiers, and locations, and redacts them while leaving symptom descriptions intact. The central claim is that this selective anonymization balances utility and privacy, and that its feasibility is demonstrated by the PII detector's 89.64% accuracy on IMCS21, one of three datasets evaluated. The interview study with 12 users supplies the motivating finding: users want anonymity and control, but platforms currently offload the responsibility of protecting privacy onto users.
Load-bearing premise
The reported 89.64% accuracy assumes the LLM-generated labels used as ground truth are correct; if those labels are systematically wrong, the accuracy measures agreement with a biased reference rather than true detection.
Editorial extensions
If this is right
- Patients can review a redacted version of their consultation before sending it, so names, contact details, and other identifiers stay off the platform's servers.
- The redaction is selective: symptom descriptions and medical content remain intact, which preserves the consultation's usefulness for the doctor.
- Because the model runs locally, the raw consultation text does not need to be transmitted for the redaction step itself.
- A relatively small, 4-billion-parameter model appears sufficient for accurate PII detection on this medical dataset, supporting the idea of running the tool on consumer devices.
Reading between the lines
- If the reported accuracy reproduces under human-verified labels, the same local-redaction pattern could generalize to other sensitive text domains, such as mental-health support chats or legal advice.
- SafeShare's 'privacy labor' framing points to a product direction in which platforms handle privacy filtering on the client side rather than expecting patients to self-censor, a shift from burden to user agency.
- A natural next step, already hinted at in the manuscript, is to re-run the evaluation with labels produced by human physicians, which would test how far the reported accuracy transfers to human-agreed ground truth.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports a qualitative interview study (N=12) of privacy perceptions in online medical consultations, identifies what it calls the 'privacy labor' burden placed on users, and presents SafeShare, a localized-LLM interaction technique that redacts personally identifying information in real time. The abstract claims SafeShare 'balances utility and privacy through selectively anonymizing private information' and reports a single headline result: 89.64% accuracy for the PII detection module with Qwen3-4B on the IMCS21 dataset, with two other datasets mentioned but not numerically reported. The methods excerpt also discloses that the ground-truth labels for this evaluation were produced by an advanced LLM rather than human annotators, with Cohen's kappa of 0.81 against manual coding, and the authors note that future work could benefit from human physician annotation.
Significance. If substantiated, the core idea is valuable and timely: a local, real-time redaction tool could give patients more agency over what is shared in online consultations, and the qualitative finding about 'privacy labor' is a plausible and potentially useful framing. The authors deserve credit for being transparent about the LLM-generated annotation step and for reporting Cohen's kappa as a partial check. However, the current evidence supports only that a prototype exists and that a detector agrees with a model-generated reference; it does not yet establish the central 'balances utility and privacy' claim, because no precision/recall, error bars, baselines, or downstream utility evaluation are reported.
major comments (3)
- [Abstract and Technical Evaluation] The central claim that SafeShare 'balances utility and privacy through selectively anonymizing private information' is not supported by the reported evidence. The only quantitative support is a single accuracy figure (89.64%) on IMCS21, with no error bars, no precision/recall at the entity or token level, no per-category breakdown, no comparison against a baseline (e.g., a conventional NER tagger or a non-fine-tuned general-purpose LLM), and no assessment of whether the redacted consultations remain useful for patients or clinicians. Please provide a fuller evaluation: category-level metrics, a confusion analysis, a baseline comparison, and at least a small human rating of the utility of redacted consultations.
- [Methods, annotation step] The ground-truth labels for the main accuracy result were generated by 'an advanced LLM' rather than human annotators, while the evaluated detector (Qwen3-4B) is also an LLM. As reported, the 89.64% figure measures agreement with a model-generated reference, not correctness against true PII. The reported Cohen's kappa of 0.81 against manual coding is a useful partial check, but it does not establish per-category detection performance, and systematic disagreements between the annotator model and human judgment could shift the headline number. Please report accuracy and kappa computed on the human-annotated subset, provide per-category precision and recall, and include an error analysis of the disagreements between the annotator LLM and manual coding.
- [Datasets and results reporting] The abstract says the evaluation covers three datasets, but the manuscript excerpt reports results for only one dataset (IMCS21). The other two datasets and their results are not described, and the choice of Qwen3-4B as the sole model is not justified. For the reported claim to be load-bearing, the paper should state the size and class distribution of each dataset, the inference settings, and the results for all three datasets, or explicitly restrict the claim to the single dataset that was actually evaluated.
minor comments (4)
- [Abstract] Please fix the grammatical and typographical issues: 'on 3 dataset' should be 'on three datasets', and 'through selectively anonymize' should be 'by selectively anonymizing'.
- [CCS Concepts and front matter] The CCS Concepts and copyright lines contain garbled placeholder glyphs in the provided manuscript; the camera-ready version must use the correct ACM template rendering.
- [Technical evaluation details] The evaluation section should specify the number of test instances, the prompt template used for Qwen3-4B, the decoding parameters, and the exact definition of accuracy (e.g., token-level, entity-level, or conversation-level), since none of these details appear in the excerpt.
- [Qualitative study reporting] The interview study would be easier to assess with a description of the interview protocol, participant demographics, recruitment procedure, and coding scheme; the current excerpt reports only high-level themes without showing how they were derived.
Circularity Check
No significant circularity; the evaluation's LLM-generated gold labels are a validity limitation, not a by-construction reduction.
full rationale
The paper's central claim is an empirical evaluation of SafeShare's PII detection module (89.64% accuracy with Qwen3-4B on IMCS21). The derivation chain is: interviews about privacy perceptions, design of SafeShare, then a technical evaluation on three datasets. The evaluation uses gold labels produced by an advanced LLM, with Cohen's kappa 0.81 against manual coding. This is a genuine threat to construct validity because the reported accuracy measures agreement with an LLM annotator rather than with independently verified ground truth, and the paper itself concedes that 'future work could benefit from having human physicians perform the annotation task.' However, this is not circularity in the defined sense: the accuracy value is not forced by construction, the detector is not shown to be the same model that generated the labels, and the kappa 0.81 provides partial independent grounding against human judgment. No load-bearing self-citation, no imported uniqueness theorem, and no fitted-parameter-renamed-as-prediction appear in the manuscript. The LLM-annotation issue is a correctness/validity risk, not a circular-derivation step, so the circularity score is 0.
Assumptions & free parameters
assumptions (4)
- domain assumption The 12 interview participants sufficiently represent online medical consultation users.
- ad hoc to paper LLM-generated labels are a valid ground truth for PII detection in medical consultations.
- ad hoc to paper Accuracy is an adequate measure of PII detection efficacy.
- domain assumption Real-time redaction preserves the clinical utility of consultation messages.
Cite this review
Pith. "Pith review of From Patient Burdens to User Agency: Designing for Real-Time Protection Support in Online Health Consultations." pith.science (2026). https://pith.science/paper/4C7SWM3W
@misc{pith2026250800328,
author = {Pith},
title = {Pith review of: From Patient Burdens to User Agency: Designing for Real-Time Protection Support in Online Health Consultations},
year = {2026},
howpublished = {\url{https://pith.science/paper/4C7SWM3W}},
note = {Machine review of arXiv:2508.00328}
}
read the original abstract
Online medical consultation platforms, while convenient, are undermined by significant privacy risks that erode user trust. We first conducted in-depth semi-structured interviews with 12 users to understand their perceptions of security and privacy landscapes on online medical consultation platforms, as well as their practices, challenges and expectation. Our analysis reveals a critical disconnect between users' desires for anonymity and control, and platform realities that offload the responsibility of ``privacy labor''. To bridge this gap, we present SafeShare, an interaction technique that leverages localized LLM to redact consultations in real-time. SafeShare balances utility and privacy through selectively anonymize private information. A technical evaluation of SafeShare's core PII detection module on 3 dataset demonstrates high efficacy, achieving 89.64\% accuracy with Qwen3-4B on IMCS21 dataset.
Forward citations
Cited by 1 Pith paper
-
Understanding Users' Privacy Perceptions Towards LLM's RAG-based Memory
Users of LLM chatbots hold incomplete, often mistaken mental models of memory features, yet actively trade privacy against personalization and demand granular control and transparency over how memories are stored, use...
Reference graph
Works this paper leans on
-
[1]
privacy labor
From Patient Burdens to User Agency: Designing for Real-Time Protection Support in Online Health Consultations Shuning Zhang zsn23@mails.tsinghua.edu.cn Tsinghua University Beijing, China Ying Ma School of Computing and Information Systems University of Melbourne Melbourne, Australia ying.ma1@student.unimelb.edu.au Yongquan ‘Owen’ Hu yongquan@ahlab.org Au...
2025
-
[2025]
In �ACM, New York, NY, USA, 9 pages
From Patient Burdens to User Agency: Designing for Real-Time Protection Support in Online Health Consultations. In �ACM, New York, NY, USA, 9 pages. https://doi.org/10.1145/nnnnnnn.nnnnnnn 1 Introduction Online medical consultation [2], a service allowing patients to seek remote advice from healthcare professionals, has seen explosive growth, especially d...
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.