We secured our own brand TLD, .reputationpartners. Your brand should too.See how
  1. Reputation Partners
  2. Journal
  3. Reviews

17 min readView as Markdown

Respond to Patient Reviews Without Breaching Privacy

How to respond to patient reviews without breaching HIPAA or UK and Australian rules: what to say, what never to confirm, and reply wording that is safe.

You can respond to a patient review without breaching privacy rules only if the reply says nothing that reveals who the reviewer is, whether they were a patient, or anything about their care. Under the US HIPAA Privacy Rule, the UK confidentiality and data protection framework and the Australian health privacy and advertising rules, the safe public reply is a short, general one that points the reviewer to a private channel.

The reviewer's own post does not change that. A patient can say whatever they like about their care in public. The clinic that treated them cannot confirm, correct or answer any of it using what it knows from the chart, the diary or the front desk. This guide sets out what each jurisdiction says, a reply structure you can reuse, and the awkward cases. It is information, not legal advice, and your own counsel or defense organization decides how it applies to your clinic.

The short version
  • A public reply that confirms the reviewer was a patient, or adds any detail about a visit, can itself be a privacy breach, whoever posted first.
  • The safe reply is general: thank them, state your standard practice, give a private contact route, and say nothing that links the reviewer to your clinic.
  • The US, UK and Australia agree on the core rule and differ on the edges: US enforcement is documented in settlements, UK guidance is about explicit consent, and Australia adds a ban on testimonials in advertising that can reach your replies.
  • Positive clinical reviews are a trap too: quoting or answering them on pages you control can breach Australian advertising law and, in the US, needs patient authorization.
  • Use a fixed approval step, a locked set of reply templates, and a rule that nobody answers from memory of the patient.

A public reply is a disclosure, even when it is polite

A public reply to a review is a disclosure of information to everyone who reads it, so it is judged by the same rules as any other disclosure.

The mistake has a common shape: a manager writes a reply that is accurate, calm and specific, and the specificity is the breach. "We offered you an earlier slot on the 14th and you did not attend" confirms that the author was a patient, that they had an appointment, and that they missed it. Each of those is information the clinic holds only because of the care relationship.

Related readBought a Business With Bad Reviews: What to Do

Two ideas run through everything below. Confirming that someone is a patient is itself information about them, before any detail is added. And silence is a legitimate reply: the Royal Australian College of General Practitioners fact sheet on online reviews says that doing nothing is generally the recommended starting point, and UK indemnity guidance from MDDUS also lists not responding as an option.

What HIPAA says about replying to a review

HIPAA allows a covered clinic to disclose protected health information in a review reply only if the Privacy Rule permits that disclosure, and the summary lists no exception for answering criticism. The HHS summary of the Privacy Rule says a covered entity must obtain the individual's written authorization for any use or disclosure that the Rule does not otherwise permit or require. Its list of permitted disclosures covers things such as treatment, payment, health care operations and certain public interest purposes. Defending a clinic's reputation in public is not on it.

Two definitions matter. First, who is covered: health plans, health care clearinghouses, and any health care provider that transmits health information electronically in connection with standard transactions such as claims. Second, what is protected: individually identifiable information about a person's past, present or future physical or mental health, the care they received, or payment for it. The information is protected if it identifies the person or there is a reasonable basis to believe it could. The summary names names, addresses and birth dates as common identifiers.

That reaches further than most staff expect. The protected information is not only the diagnosis. A reply that says "we were glad to see you for your appointment" tells readers that the named reviewer attended a health service. For some specialties that fact alone is sensitive. The summary also sets a minimum necessary standard, and a rebuttal of a review can meet its purpose with no disclosure at all.

Related readGoogle Reviews Disappeared: Why and What to Do

Penalties are civil money penalties set by tiers that depend on whether the entity knew or should have known; the HHS page lists inflation-adjusted figures that are revised over time, so check the current schedule before quoting one. Whether the Rule applies to you depends on covered-entity status, so confirm that with counsel if you are unsure.

What the US settlements show

The enforcement record on review replies is documented in HHS resolution agreements, and it is consistent: the Office for Civil Rights treats disclosing patient information in a public reply as an impermissible disclosure, and it has settled with individual practices.

$10,000Elite Dental Associates, Texas, settlement announced October 2, 2019
$23,000New Vision Dental, California, settlement published December 14, 2022
$30,000Manasa Health Center, New Jersey, settlement announced June 5, 2023

Amounts and dates are from the HHS Office for Civil Rights resolution agreement pages and press release read for this guide; each practice also agreed to a corrective action plan.

The Elite Dental page describes social media disclosures of patients' protected health information but not the specific posts, so this guide draws no further detail from it.

The New Vision Dental page says the practice disclosed patient information in its responses to online patient reviews on social media, and states, "This practice is illegal under HIPAA."

The Manasa case is the most useful because the facts are plainer. Manasa is a New Jersey provider of adult and child psychiatric services. OCR received the complaint in April 2020. The press release says the practice posted a response to a patient's negative online review that disclosed the patient's information, and that it had also failed to maintain Privacy Rule policies and procedures for PHI. Manasa paid $30,000 and agreed to a corrective action plan that OCR monitored for two years. OCR's director said of the practice's conduct, "Simply put, this is not allowed."

The second Manasa finding, the missing policies, shows that regulators look at the process around the reply as well as the reply.

Related readRemove a Clutch Review: What Agencies Can Dispute

What UK guidance says: confidentiality first, data protection beneath it

In the UK there is no single rule about review replies. The position is built from professional confidentiality duties and data protection law, and the practical advice from the regulator-adjacent sources is the same as in the US: do not confirm, do not detail, take it offline.

The General Medical Council has guidance titled "Confidentiality: responding to criticism in the media". It took effect on 25 April 2017 and was revised on 13 December 2024. The guidance says it applies to social media as well as print and broadcast. The page we read does not mention online review sites by name, and it does not address a patient's own public post, so applying it to a review reply is a reasonable reading rather than a stated rule. Its content is clear:

  • A doctor must not put information learned in confidence about a patient into the public domain without that patient's explicit consent.
  • A doctor should usually limit a public response to an explanation of the legal and professional duty of confidentiality.
  • When denying an allegation, a doctor must be careful not to reveal personal information beyond what a simple denial needs, and the guidance extends that to what is revealed "by omission or inference".

That last point closes the loophole in most drafted replies. "We have no record of any such incident" tells readers that a record exists and was searched.

On the data protection side, the Information Commissioner's Office lists data concerning health among the special categories, meaning personal data related to the physical or mental health of a person. Its page says that processing special category data requires an Article 9 condition, and that inferred data can be special category data too. A reply naming the reviewer and the service they used can reveal health status, so we treat it as special category processing. Put the question of whether any Article 9 condition could cover a public rebuttal to your data protection officer or solicitor.

Related readGoogle Seller Ratings: Remove a Review or Fix the Stars

For dental practices, MDDUS, a UK indemnity organization, advises in its guidance on responding to online practice criticism that if you reply publicly you keep it brief, give no detail about the poster, and do not confirm they are a patient, then invite them to contact the practice directly or use the complaints policy. It also cites the General Dental Council's standards: a dentist must not make personal, inaccurate or derogatory comments about patients or colleagues, including on social media. MDDUS advises against switching comments off as a way to avoid the problem. That is an indemnifier's advice, not a regulator's rule.

UK advertising is a separate question from confidentiality. The GMC's social media guidance says doctors who promote services must comply with the rules of the Committee of Advertising Practice, the Advertising Standards Authority (ASA) and the Competition and Markets Authority, and it does not address patient reviews specifically. The ASA's health advice says a marketer using a testimonial must hold evidence that it is genuine and have the speaker's permission, and factual efficacy claims in it need evidence. We found no UK rule that mirrors Australia's testimonial ban, so treat clinical praise on your own pages as a claim you must be able to substantiate.

What Australia adds: privacy, and a ban on testimonials in advertising

Australia has two overlapping regimes, and the second one is unique among the three. Privacy law restricts what you say. Health advertising law restricts what you do with praise.

On privacy, the Office of the Australian Information Commissioner says any organization that provides a health service and holds health information is covered by the Privacy Act, whether or not it is a small business, and that health information is among the most sensitive types of personal information. The Royal Australian College of General Practitioners fact sheet puts the working rule bluntly: you must not breach a patient's privacy no matter what they have posted themselves. It adds that you should not confirm whether a reviewer attended the practice, should not assume that a person who contacts you directly is the actual patient, and should avoid responding to reviews that discuss clinical issues.

Related readRemove a RateMDs Review, and Vitals or WebMD Care

The Australian Health Practitioner Regulation Agency (Ahpra) adds a point about identification. Its social media guidance warns that individual details may not breach confidentiality on their own, but combined online information could identify a patient or someone close to them. The same page says the National Law, the Code of conduct and the advertising guidelines apply to social media as they do in person.

The advertising rule is the distinctive one. Section 133(1)(c) of the National Law prohibits advertising a regulated health service in a way that uses testimonials or purported testimonials. Ahpra's testimonial guidance, as of the version read, defines a testimonial as a recommendation or positive statement about a clinical aspect of a service, such as the symptom, diagnosis, treatment or outcome. Comments about customer service or communication style are not testimonials. Patients remain free to post reviews.

Where this meets replies, the guidance is careful but not comfortable:

  • On pages and social media accounts you control, you are responsible for keeping testimonials off them, even if the platform does not let you edit or remove reviews. Disabling the review function is one way to comply.
  • On third-party sites you do not control, you are not required to remove reviews. But Ahpra's advertising guidelines say a provider "should take care" if they choose to engage with reviews there, because doing so may be treated as using a testimonial to advertise.
  • Editing or selecting reviews, such as removing negative ones or publishing only some, can be false or misleading.

A warm reply to a glowing clinical review ("so glad your knee surgery went well") breaches privacy and may also engage the advertising rule. The guidelines read for this guide took effect on 14 December 2020; check Ahpra's advertising hub for the current version before you rely on them.

How the three regimes compare on a review reply

The three agree on the rule and differ in the enforcement route and the extra trap. This table is the version to give to a practice manager.

QuestionUnited StatesUnited KingdomAustralia
Source of the dutyHIPAA Privacy Rule for covered entitiesProfessional confidentiality duties plus UK GDPRPrivacy Act plus the National Law and Code of conduct
Confirm the reviewer was a patient?No, it is identifying PHINo, per indemnity and confidentiality guidanceNo, per the RACGP fact sheet
Does the reviewer posting first change it?The summary lists no exception for itExplicit consent is the test; the GMC page does not address itRACGP says no
Documented enforcementThree HHS settlements, 2019 to 2023No settlement on review replies found in sources readNo review-reply case found in sources read
Extra trapAuthorization needed to quote a patientSpecial category data rulesTestimonial ban in advertising

The two "none found" cells are limits of the sources read, not a statement that no action has been taken.

A reply structure that discloses nothing

A safe reply has four parts in a fixed order, and the whole reply should fit in three or four sentences. Every extra clause is a chance to confirm something.

Related readAI-Generated Fake Reviews and the FTC Rule
Build a reply in four moves
  1. Thank without acknowledgingThank the reviewer for the feedback in general terms. Do not use their name, do not write "your visit" or "your appointment", and do not refer to what they said happened.
  2. State your standard practiceSay what the clinic always does, as a general standard: "We take every concern seriously and review it through our complaints process." This answers the allegation without confirming the event.
  3. Give a private routeOffer a contact route that is the clinic's general line or complaints address, with hours. Do not ask for their name, date of visit or condition in the public thread.
  4. StopDo not add a second paragraph, a defense of the clinician, or a request to update the review. Do not reply again to the same thread unless a new, general point needs making.

A reply built this way reads as follows, with no confirmation in it:

Thank you for taking the time to share feedback. We take every concern seriously and handle it through our complaints process, which protects the privacy of everyone involved. Please contact our practice manager on the main line during opening hours so we can look into this properly.

Notice what it avoids: the words "patient", "you attended" and "your treatment". The Manasa case is the cautionary example of a reply that engaged with the substance.

Cases that break the template

Decide the handling of unusual reviews in advance. In each case your reply may respond to the reviewer's request, never to their medical facts.

The reviewer posts their own health details

The reviewer naming their diagnosis or treatment does not permit you to discuss it. Reply with the standard general message and the private route. Do not repeat the detail, even to deny it. The RACGP fact sheet advises avoiding any response to reviews that discuss clinical issues, and the GMC guidance applies the same caution to denials.

The reviewer is not a patient, or you cannot tell

Reply in the same general words. The RACGP fact sheet warns against assuming a person contacting you directly is the actual patient, so identity checks belong in the private channel, using your normal verification process, not in the public thread. Saying "we cannot find anyone by that name" is a disclosure in the other direction, because it tells readers you searched.

The review is praise about clinical results

Leave it alone on pages you control and think twice before replying on third-party sites. In Australia, clinical praise is a testimonial and a reply that amplifies it can be treated as advertising. In the US, thanking a reviewer for a successful outcome confirms they were treated. A general "thank you for the kind words" with no detail is the most that sources read support.

The review names a clinician or staff member

Do not defend the individual in public. Staff conduct is an internal matter, and any reply that explains what the person did or did not do discloses the care. Route it to the complaints process and tell the staff member not to comment publicly.

Related readCheckatrade Review Dispute: What Can Be Removed

The review is abusive, threatening or plainly fake

The reply is not the right tool. Report it to the platform under its content policies and keep a record. MDDUS advises acting quickly on abusive or threatening posts, asking the platform to remove them, and contacting the police about serious threats. Do not publish the reviewer's identity as part of the response.

Who replies, who approves, and how fast

The Manasa finding about missing policies tells you what to build: a small process, written down, with one owner. Nothing in the sources read sets a deadline for replying, so the timings below are a working pattern, not a legal requirement.

WhenActionWhoWhy it matters
Day 0Log the review, take a screenshot, do not replyReception or marketingStops same-day replies written in anger
Day 1Practice manager classifies it: praise, complaint, clinical, abusivePractice managerFixes the template to use
Day 1 to 2Post the approved general reply, or decide to stay silentOne named account holderKeeps posting rights narrow
Within the weekHandle the underlying complaint through the private routePractice manager and clinicianResolves it where it can be discussed
MonthlyReview the log for themes and for any reply that went out unapprovedPractice managerShows the policy is operating

Three controls do most of the work: limit posting rights to one or two named people, keep locked templates with departures treated as escalations, and keep the treating clinician out of the reply box. Treat an agency or an AI writing tool the same way: never paste a review and the matching chart note into one to get a draft.

The next step if a reply is already live or a review is waiting

If a reply you suspect is a breach is already public, the first move is to take advice, not to post again. A second reply to apologize for the first usually restates what was disclosed.

If this is happening to you now

Do not reply again until you have decided who owns the response, taken the screenshot, and checked the draft against the four moves above. If you want a second pair of eyes on how a review, a reply and the clinic's search results fit together, a private audit is the place to start. Request a private audit.

Reputation Partners works on how reviews and replies appear in search engines and AI answers. We do not give legal advice: whether a disclosure has already occurred is a question for your counsel and your regulator.

Questions people ask

Can I reply if the patient posted their own details first?

You can reply, but only in general terms. The patient's choice to post does not release your clinic from its duty. The RACGP fact sheet says you must not breach a patient's privacy no matter what they posted themselves, and the HHS Privacy Rule summary lists no exception for answering a patient who spoke first. Use the standard reply and the private route.

Related readOpenTable Review Removal and Resy: What Venues Can Do

Is thanking a reviewer by name a breach?

Using the name is the riskiest part, because it ties the review to a person you may be treating. A reply that repeats the name confirms to every reader that a named person may have used your service, whatever name the reviewer chose to post under. Thank them without the name and without any words that suggest a visit took place.

Can I say "we have no record of this person"?

No. That sentence discloses that you searched your records and tells readers something about the person's relationship to your clinic. The GMC guidance warns against revealing information "by omission or inference". Say instead that you take all concerns seriously and give the private route.

Do these rules apply to a small practice or a single clinician?

HIPAA applies to covered entities, which includes providers that transmit health information electronically in standard transactions, and the settlements above involved individual practices. In Australia the OAIC says any organization providing a health service and holding health information is covered. In the UK, professional confidentiality duties attach to the individual. Check your own status with counsel.

Can I use a patient's positive review on my website?

In the United States, quoting a patient generally needs their written authorization under the Privacy Rule: HHS settled for $25,000 with a physical therapy practice in 2016 over testimonials posted without valid authorizations. In Australia, praise of a clinical aspect is a testimonial and cannot be used on pages you control; non-clinical comments are not. In the UK, the ASA expects proof the review is genuine and evidence for any efficacy claim.

Should I ask the platform to remove a negative review I cannot answer?

Only if it breaks the platform's rules, such as a breach of its posting policy or defamation, as the RACGP fact sheet puts it. Most sites will not remove a review just because you disagree with it. Use the platform's report route and keep a record.

Reputation Partners desk

Notes from the desk. Written by the people who handle the cases.

Tell us what Google says. Quietly.

A partner reads every request personally. Every engagement is quoted individually, after an audit.