Version 3.8 · Effective October 13, 2026
Trial Friend is a rare disease information hub for patients and caregivers, offering disease guides, drug information, treatment timelines, comparison charts, clinical trial search, and an AI chat assistant. We take your privacy seriously, especially given the sensitive nature of health-related searches. This policy describes what data we collect, how we use it, how long we keep it, and what rights you have over it.
These are the short versions of the commitments in this policy, in plain words.
Zero ad trackers. No Google Analytics, no Facebook pixel, no retargeting. Ads won't follow you around the internet because of what you read here. You can check this yourself with The Markup's Blacklight scanner.
We never sell your data. We don't share it with advertisers or drugmakers either. Sponsors see only total counts, such as how many people viewed a page.
No account needed. You can search, read and use every tool without signing up, and we never store a full ZIP code.
What we do collect. Visit counts from Plausible, which uses no cookies and stores no IP addresses, and from our own counter, which may save a random ID in your browser and notes which pages and tools you use and some of what you choose in them, such as a condition or an age range. Your email, only if you sign up for alerts. Chat questions, kept for 90 days under a random ID so we can check our answers, with replies written by Anthropic's AI. The details are below.
If you sign up for trial alerts, we collect your email address and the condition you are interested in. We use this only to send you notifications about new matching trials and, if you opt in, educational content and disease updates. We never sell, share, or rent your email address to third parties, including pharmaceutical companies. An email address tied to a rare disease condition is sensitive data, and we treat it with the same care we would treat health information, even though it is not technically protected health information (PHI) under HIPAA.
Every time you sign up, change a preference, unsubscribe, or request deletion, we log that event in an audit trail. Each entry records the action, the time, the condition and type of email it applies to, the version of this policy in effect, and a one-way hash of your email. Until you ask us to delete your data, the entry also holds your email address, IP address and browser type, even after you unsubscribe. We keep these records while you are subscribed and for up to 7 years after your last change, for legal compliance, and then delete them automatically. When you ask us to delete your data, we erase the email address, IP address and browser type from these entries and keep only the hash. This protects both you and us by documenting that we have your permission to send you email and that we honored any request to stop.
When you use the AI chat assistant, your messages are sent to our server and processed using Anthropic's Claude API to generate responses. Your messages go to Anthropic as you type them, along with the page you are on. Each conversation is stored in our database (Supabase) with a random chat ID, the same random browser ID our analytics use, your approximate city and region, and your browser type. It is never linked to your email address. Email addresses, phone numbers and Social Security numbers are removed from your messages before a chat is saved, but they still reach Anthropic as typed, so please don't share personal details. To answer, the assistant may search public databases, such as ClinicalTrials.gov and those of the FDA, NIH, CDC and SEC, using terms from your question. Chat logs are automatically deleted after 90 days.
If you ask us to email you a copy of a chat, your email address and the conversation are sent through our email provider, Resend, to deliver it. We do not save that address.
If you use “Something wrong? Tell us” under a chat answer, your note, the question and the answer are emailed to our team so we can check it. The form does not ask for your email address, and reports are not saved on our website or database.
Trial Friend uses two analytics systems, and neither is an ad tracker. Google Analytics was removed on October 12, 2026. Plausible Analytics counts page visits without cookies. It records the page address, the site that sent you, your browser and device type, and an approximate location worked out from your IP address. Plausible does not store IP addresses and cannot recognize the same visitor from one day to the next; its data is stored in the European Union. We also use a custom first-party analytics system (powered by Supabase) to count page views and feature usage on our own infrastructure. It records the pages you view, the site that sent you, your approximate city, region and country, your browser type, and what you do on the site, such as trial clicks, the condition you search for and, in Match Me, an age range, the sex you chose and whether and how many treatments you take. It never records your exact age, a full ZIP code, medicine names or your IP address. For visitors in the European Economic Area, the UK, Switzerland and Washington state, it keeps no identifier on your device until you accept: each visit is counted on its own and is forgotten when you leave. Neither analytics system records your email address.
To power weekly change digests, we store snapshots of publicly available trial data from ClinicalTrials.gov, including trial titles, statuses, site locations, enrollment targets, and eligibility criteria. This is public clinical trial data, not personal information. It is used solely to detect changes and notify subscribers who are tracking those conditions.
Trial Friend may license reports and datasets built from public clinical trial records, such as changes to trial status, eligibility and site lists over time. These products contain only public trial information and our analysis of it. They never include information about visitors, subscribers or chat users. Any report we share about how the site is used contains only totals, such as visits or actions per condition, never individual visits, IDs or locations.
When you type a condition into a search box, your text is first matched against our own disease list in your browser. When you run a trial search on the homepage, the condition and location you enter are sent from your browser to ClinicalTrials.gov (a U.S. government service) to fetch matching trials; a 5-digit ZIP code is first turned into map coordinates, as described under Geocoding below. When you visit a disease page, your browser fetches live trial data from ClinicalTrials.gov using the condition name from our database. When you use Match Me, the condition you select is sent to ClinicalTrials.gov to fetch its open trials. The matching happens in your browser, and your exact age, ZIP code and medicine names stay there; our analytics keep only the coarse answers described under Usage analytics. ClinicalTrials.gov's privacy policy applies to what it receives.
When you use the Symptom Finder, the symptoms you pick are sent to our server, which looks them up one at a time in Orphanet's public rare disease database. We keep no record of who searched. When you use gene search, your browser looks up the gene you choose directly in Orphanet's public database (api.orphadata.com).
When you use the alpha-gal checker, each medicine you look up, and the text you type while it suggests names, is sent to our server, which checks it against FDA labels and the National Library of Medicine's drug database (RxNav). We count lookups by medicine name, with no ID attached.
The “Translate to plain English” feature rewrites a trial's eligibility criteria in your browser using built-in rules. The text is not sent to our server or to any AI model.
Some drug pages include an interactive treatment timeline that asks you when you started a medication, which formulation you take, whether you have had follow-up labs, and which side effects you are experiencing. This information is used solely to personalize the timeline shown on that page. It is stored in your browser's localStorage so your progress is saved if you return to the same drug page. This data is never sent to our servers, never stored in any database, and never included in analytics. You can clear your Treatment Arc data at any time using the reset option on the timeline or by clearing your browser's localStorage.
Trial Friend may occasionally offer voluntary surveys, such as patient experience questionnaires, quality-of-life assessments, or feedback forms. Participation is always optional. Survey responses are stored without your name or email unless you choose to provide your email for follow-up. Total, de-identified survey results may be shared with patient advocacy organizations or research partners to help improve rare disease care, but individual responses are never sold or shared in identifiable form. You will always be told at the start of a survey how your responses will be used.
We do not ask for your name, diagnosis, medical history or medications, and we do not collect protected health information (PHI) as defined by HIPAA. We do not use cookies for advertising or remarketing. We do not build behavioral profiles, and we do not sell, rent or share personal data, or information about what you read or search on Trial Friend, with third parties. Unlike platforms such as Facebook or TikTok, Trial Friend does not track you across websites, build advertising profiles based on your health searches, or sell your data to brokers. If we display an ad, the advertiser knows nothing about you. They see aggregate numbers (how many people viewed the ad), not who you are or what you searched for. While we do collect email addresses tied to rare disease conditions (which is sensitive even if not technically PHI), we do not collect or store any clinical or medical details about you beyond the condition you select when signing up and the coarse tool answers described above. Treatment timeline responses are saved in your browser's localStorage and are never transmitted to us.
Trial Friend sets no advertising, remarketing or analytics cookies, and Plausible uses none. Our custom Supabase analytics system stores a random visit ID in your browser's localStorage (tf_device_id) and a few visit details in sessionStorage so that pages in one visit, and later visits, are counted as the same browser; this ID is never linked to your email. A short-lived tf_cc cookie holds only your country code, and for Washington visitors a tf_rg cookie holds the state code, so we know whether to ask for consent. We use localStorage to remember your consent choice so we do not ask again. Treatment timeline data, your saved alpha-gal checker list and its sensitivity settings are stored in your browser's localStorage so your progress persists between visits; they are never sent to our servers. The chat keeps a random chat ID in sessionStorage while the tab is open. For visitors in the European Economic Area, the UK, Switzerland and Washington state, the analytics visit ID is only stored after you accept our consent banner, and anyone can turn it off on our Consumer Health Data Privacy Policy page. You can decline, or clear these items from your browser at any time.
Email addresses: Retained until you unsubscribe or request deletion.
Chat logs: Automatically deleted after 90 days.
Consent records: Retained while you are subscribed and for up to 7 years after your last change, for legal compliance, then deleted automatically. After you ask us to delete your data, each consent record keeps only a one-way hash of your email, not the address, your IP address or your browser type. A one-way hash cannot be turned back into your address, but it would match the same address if it were entered again, which is how we keep honoring a request to stop.
Usage data: Our own analytics records, each with a random browser ID and approximate location but never your email address, are kept for as long as we use them to show how the site changes over the years. If we stop needing that detail, we will keep only totals.
Trial Friend may send you the following types of email, depending on your preferences:
Trial Alerts (on by default): Notifications when new clinical trials open for your condition, and summaries when existing trials change status, add new sites, update enrollment targets, or revise eligibility criteria. We check ClinicalTrials.gov daily for new trials and re-check each condition's trials for changes about every 9 days.
Disease Updates & Education (on by default): Occasional updates about your condition, such as new treatments, FDA approvals, and plain-language research explainers. Most months that means 1 or 2 emails at most, though news of an FDA decision for your condition is always sent, even in a month that has already had 2.
Product Improvement (on by default): Records your preference only. Site analytics are controlled by the consent setting on our Consumer Health Data Privacy Policy page, and they never record your email address.
Foundation Events & Advocacy (opt-in, off by default): Notifications about rare disease events, conferences, and advocacy opportunities related to your condition. These messages may be written on behalf of patient advocacy organizations or foundations, but Trial Friend sends them directly from our own email infrastructure. We never share your email address with the organization. They provide the content; we deliver it. The foundation never sees your email or any identifying information about you.
Partner Communications (opt-in, off by default): Occasional messages from trusted organizations working in your condition area, delivered by Trial Friend on their behalf. As with foundation events, your email is never shared with the partner. All messages are sent through our system, and partners receive only aggregate statistics (like how many emails were delivered). You can turn this off at any time.
Additional categories (patient research surveys, product updates, and patient advisory opportunities) may be added in the future. These will always be opt-in and off by default.
You can manage all email preferences at any time from your preferences page (a personalized link is included in every email we send) or by emailing jason@trialfriend.com.
You have the following rights regarding your data:
Access: Email us for a copy of the data we hold about you.
Correction: Update your email address or preferences.
Deletion: Request removal of all your data. We will permanently delete your email address, condition and subscriptions, and the record of which emails we sent you, within 30 days. For legal compliance we keep only a one-way hash of your email, with the dates and choices in your consent records.
Restriction: Turn off any email category, or unsubscribe, without deleting your data.
Portability: Email us and we will send your data in a standard format, such as CSV or JSON.
Objection: Opt out of any processing category at any time.
To exercise any of these rights, visit your preferences page or email jason@trialfriend.com. Unsubscribing stops all emails but keeps your record so we know not to contact you. Requesting deletion removes your email address and subscriptions as described above.
We do not sell or share your personal information as defined by the California Consumer Privacy Act (CCPA) and the California Privacy Rights Act (CPRA). You can contact us at jason@trialfriend.com to exercise your California privacy rights, or visit our Do Not Sell or Share page.
Our Consumer Health Data Privacy Policy explains what health-related data we collect, why, who processes it, and how to access, delete or stop it, including how to appeal a decision, under Washington's My Health My Data Act. Visitors in Washington are asked before our analytics saves an ID in their browser, and anyone can turn that off on the same page.
Trial Friend is operated in the United States. If you access Trial Friend from outside the U.S., your information may be transferred to and processed in the U.S. By using Trial Friend, you consent to this transfer. We build our consent and deletion infrastructure to meet the standards expected by international privacy regulations, even where we are not formally required to comply with them.
Trial Friend may be supported by sponsors, shown as a credit such as “Supported by” followed by the organization's name, and by clearly labeled advertising. Sponsors and advertisers receive only aggregate numbers, such as how many times a page or an ad was viewed. They never receive your email address, your condition preferences, your chat messages, your ZIP code, or any record of which pages you personally visited, and no data about you is used to target ads. The rules that keep sponsors away from editorial content are on our editorial standards page.
Trial Friend may display clearly labeled advertisements, sponsored content, or downloadable resources from pharmaceutical companies, patient advocacy foundations, research organizations, or other partners. All paid placements are visually distinct from editorial content and marked as “Sponsored” or “Advertisement.” Sponsors and advertisers have no influence over the drug information, treatment timelines, comparison charts, side effect data, or any other editorial content on Trial Friend. All medical content is independently sourced from FDA prescribing information, published clinical trials, and established medical references regardless of any advertising relationship. We do not share your email address, browsing history, or personal information with advertisers or sponsors. Advertisers may receive only totals, such as ad impressions or clicks. Trial Friend may also partner with patient advocacy foundations, research organizations, and rare disease nonprofits for educational initiatives, survey distribution, or awareness campaigns. These partnerships follow the same principle: your personal data is never shared with partners.
Plausible Analytics: We use Plausible to count page visits without cookies. Plausible does not store IP addresses or build profiles, and keeps its data in the European Union. Its data policy is at plausible.io/data-policy. We stopped using Google Analytics on October 12, 2026.
ClinicalTrials.gov: Trial data is fetched directly from the ClinicalTrials.gov API, operated by the U.S. National Library of Medicine. Their privacy policy applies to data accessed through their service.
AI processing: Chat answers are written by Anthropic's Claude API. Your messages are sent to Anthropic as you type them and stored in Supabase for 90 days, as described under Chat conversations. The plain English translation of eligibility criteria does not use AI.
Email delivery: Emails are sent through Resend, a transactional email service. Resend processes your email address solely for the purpose of delivering our emails.
Database: Subscriber data, chat logs, and first-party analytics are stored in Supabase, which runs on AWS infrastructure with encryption at rest and in transit.
Geocoding: When you enter a US ZIP code in Match Me, Trials Near Me, a disease page's trial map or the homepage trial search, your browser sends it to Zippopotam.us, a free lookup service, to get map coordinates for measuring distance to trial sites. Trial Friend never receives the full ZIP code. When one is typed into the homepage trial search, our analytics keep only its first 3 digits.
Maps and public data: Trial maps load map images from OpenStreetMap and map code from cdnjs, a Cloudflare service. Some disease pages fetch public research and drug data directly from the NIH (PubMed) and openFDA, and gene search fetches gene data from Orphanet (api.orphadata.com). Trial Friend sends these services no personal information, though, like any web request, they can see your IP address. None of them is an ad tracker.
Lookups our server makes: Our server, not your browser, looks some things up in public databases: Orphanet for the Symptom Finder, the FDA and the National Library of Medicine for the alpha-gal checker, YouTube's video search for disease names on disease pages, and the databases named under Chat conversations for the chat. These services see what is looked up, not who asked.
Partner tools: The IgA nephropathy page can show the IgA Nephropathy Foundation's specialist finder, which uses Google Analytics. It loads only if you press its “Load the specialist map” button, and the Foundation's privacy policy applies to it.
Hosting: Trial Friend is hosted on Vercel. Vercel may collect standard server logs (IP addresses, request timestamps) as part of normal web hosting. We do not use Vercel analytics or tracking features.
Trial Friend is not a covered entity under HIPAA. We do not collect, store, or transmit protected health information as defined by HIPAA. We are a search tool that helps you find publicly available clinical trial information. That said, we recognize that an email address linked to a rare disease condition is sensitive data, and we treat it with heightened care: encryption at rest, limited access, consent-based collection, and prompt deletion on request.
In the unlikely event of a data breach affecting your personal information, we will notify you by email within 72 hours of becoming aware of the breach, consistent with applicable law. We will also notify relevant authorities as required.
This policy may be updated. We will notify subscribers by email of material changes before they take effect, and we will ask for your consent before using personal information we already hold in a way this policy does not describe. The version number and effective date are listed at the top of this page. Previous versions are archived and available on request.
Questions about this policy? Email us at jason@trialfriend.com.