Spark participant app
Privacy Policy
Effective 12 November 2026 · Last updated 12 November 2026
Spark is a research instrument. You use it as a participant in a research study, with a study ID your research team gives you. Your study decides which surveys and messages you receive; a few things, listed in section 2, Spark asks of every participant whatever the study. Everything it records exists to run that research. None of it is used for advertising, sold, or shared with data brokers.
1. Who this covers
This policy describes the Spark mobile app for iOS and Android and this website at entropichealth.org, apart from Console (below). Spark is published on the app stores by Harvey Wang, who operates it under the name Entropic Health and is responsible for the data described here. Each study that runs on Spark is conducted by its own research team, under its own ethics approval; the investigators named on your consent form are responsible for that study’s conduct.
You create an account with your email address and a password. Setting up the app then asks for your waking hours and for the study ID your research team gave you, and checks with our server that a study with that ID is running; setup does not finish until one is. Study prompts are sent only to accounts enrolled in a study. If you create an account but do not join a study, we hold your email address and password, the setup choices you made (such as your waking hours), a record that check-in surveys are switched on for you, and, if you allowed notifications, the push token that lets us reach your phone along with your phone’s periodic reports on whether notifications are allowed.
Once you are enrolled, that study’s consent form and its research ethics board approval govern the research itself, including what the study may ask of you and what its investigators may do with the data. Where this policy and your signed consent form differ, your consent form governs.
Console, the web application research teams use to design and run studies (at entropichealth.org/console), is not covered here, except where this policy says what researchers can see through it.
2. What Spark collects
Your study sets what its surveys ask and which messages it sends. Everything else below applies to every participant, whatever the study, unless the item says otherwise. Anything that depends on a device permission is collected only after you grant that permission, and only for as long as it stays granted.
| Category | What it contains | Why |
|---|---|---|
| Account | Your email address and password, held in Amazon Cognito. Passwords are stored by Cognito as salted hashes and are never visible to us. Your study records are stored under a random account identifier, not under your email. Your research team can see your email address alongside them (section 6). | To sign you in and to tie your responses to one participant record. |
| Study enrollment | The study you joined, any participant number your research team assigned, and the date and time zone you enrolled in, which set where you are in the study’s schedule. Your usual number of cigarettes per day, which setup asks every participant. | To run the right study, on the right schedule, for you. |
| Study responses | Your answers to the prompts and surveys your study sends, with the times you opened and answered them, including surveys you started but did not finish. Content is set by your study and can include self-reports about mood, context, and behaviours such as cigarettes smoked or urges resisted, any text you type into an answer (such as an “Other” option), and your ratings of the messages and images your study sends you, along with which image you were shown. | These are the study’s primary measurements. |
| Location | Latitude, longitude, accuracy, altitude, speed, heading, and time. Collected only after you start location tracking and grant background location permission. Tracking is started from Spark’s location screen, which has no button in the app: your research team sends you a link to it if your study uses location. Recorded at balanced (not highest) accuracy, sent no more often than once every five minutes, and it continues while the app is closed until you stop it, withdraw Spark’s background location permission, or sign out. Nothing is recorded without that permission. | To relate where you spend your time to your study responses, as your study’s protocol describes. |
| Your surroundings | Details you enter yourself: who you live with and labels for places that matter to you. If you choose to save them, the locations of your work and school, with an address label your phone looks up from the pin you place. | To interpret your location in context, so a study can distinguish home from work rather than only knowing coordinates. |
| Health and activity | On iOS only, and only with your Apple Health permission, which Spark asks for when you open the Health data section of your profile: step count, heart rate, resting heart rate, active energy burned, exercise time, walking and running distance, stand time, and flights climbed, including readings from an Apple Watch. Each time you open or refresh that section, Spark sends today’s running totals and the most recent readings, not every sample, each with the name of the app or device that recorded it (for your iPhone or Apple Watch, the name you gave it, which often includes your own name) and Apple Health’s identifier for that source. For readings recorded by your iPhone or Apple Watch, that identifier contains a code unique to the device, which stays the same over time. Spark only reads from Apple Health. It never writes to it. | To relate physical activity and physiology to your study responses, and to show you your readings in the app. |
| Prompt delivery | Your phone’s push notification token and whether it is an iPhone or an Android phone. Which prompts were scheduled, which reached your phone and by which route, which you opened, which went unanswered, and when you deferred one. Any adjustments to your schedule. Periodic reports from your phone on whether notifications are allowed, whether its backup reminders are set, and when a notification last woke the app. | To deliver prompts at the right time, avoid sending duplicates, and let researchers distinguish a prompt you declined from one that never arrived. |
| Your settings | Your notification preferences, waking hours, time zone, and whether you have paused prompts. The app updates your time zone by itself when your phone’s changes, and each change is recorded as an adjustment to your schedule. When you first sign in, the app also records that check-in surveys are switched on for you; there is no control in the app that changes it. | To keep prompts inside your waking hours and the preferences you chose. |
| Service records | Our servers log your random account identifier, your study, your time zone, your push token and the identifiers of prompts while they handle your requests and run scheduled tasks. They record errors, which can include the prompt or schedule record involved. They do not log your answers or your email address. Your IP address is used, as your phone connects, to protect the service from abuse, and is not stored in your records. | To keep the service secure and to diagnose faults. |
Spark also keeps a copy of your data on your phone so it works without a signal. Signing out removes it, with two exceptions. First, anything you did that has not yet reached our server, such as an answer, a health reading or a change to your settings, stays on the phone, with its contents encrypted, until it can be sent. It is sent when you sign back in or, if our server turned it away, when you tap Retry. If someone else signs in on that phone instead, it is discarded unsent. Second, the phone remembers the random account identifiers of the last account to sign in and of the last to finish each setup step, and keeps a copy of your study’s settings, which name the study but not you, and the pictures from messages you were shown. Deleting the app removes Spark’s data from the phone, with two exceptions on an iPhone: the iPhone’s keychain keeps two small items (the key that encrypts unsent data, and an installation identifier used to deliver notifications), and backups of the iPhone made while Spark was installed keep their copy. Deleting the app does not delete anything from our servers (section 9).
3. Device permissions
Spark asks for these permissions and uses them for nothing else. You can decline any of them, or withdraw them later in your device settings. Declining reduces what the study can measure but does not lock you out of the app.
- Notifications
- To deliver your study prompts. Without this, prompts cannot reach you on time.
- Location, while using the app
- To show where you are on Spark’s maps, for example when you place your work or school location. It is also the first step the phone requires before background location.
- Location, in the background
- Requested only when you start location tracking on Spark’s location screen (section 2). It lets Spark record your location periodically while the app is closed. Withdrawing it ends tracking. On Android a persistent notification shows while this is running. If you have turned off Spark’s notifications, Android 13 and later hide that notification, but still list Spark among the apps running in the background, in the active-apps list at the bottom of the notification shade.
- Apple Health
- To read the activity and heart rate types listed above. You choose which types to share in the Health permission sheet, and you can change that at any time in the Health app. Spark does not ask for permission to write to Apple Health.
Spark also holds standard permissions that never ask you anything. On Android these let it connect to the internet and check whether the phone is online (including its Wi-Fi state), receive notifications, keep the phone awake briefly to handle them and vibrate for them, run background location as a visible foreground service, and set reminders again after your phone restarts. On iOS they let a notification wake the app briefly to keep your reminders in order. Spark has no access to your files or photos and cannot draw over other apps.
4. How the data is used
- To deliver your study’s prompts and interventions at the intended moments.
- To record your responses as study measurements.
- To choose, automatically, which of your study’s messages you are sent. In some periods of a study the message is drawn at random from the ones your study wrote. In others, fixed rules built into Spark choose the kind of message (about cravings, about mood, or neither), and can take into account your highest answers about craving, low mood and anxiety from the past day; one message of that kind is then drawn at random. No machine learning model is involved. Study staff can also send you a message or a check-in by hand, either one your study wrote or wording they write themselves.
- To show you your own data inside the app.
- To let the research team monitor participation and data quality, and to conduct the scientific analysis described in your consent form.
- To keep the service secure and to diagnose faults.
Our legal basis is your consent, given when you signed your study’s consent form and again each time you grant a device permission. You can withdraw either at any time.
5. What we never do
- We do not sell or rent your data, and we do not share it with data brokers.
- We do not use it for advertising, and Spark contains no ads.
- We do not track you across other companies’ apps or websites. Spark does not use advertising identifiers and contains no third-party analytics or marketing software development kits.
- We do not use your study data to train commercial machine learning models.
6. Who can see your data
Your research team. Through Console, the investigators running your study can see your email address, your enrollment details, your survey answers and message ratings, your prompt and delivery history, and whether notifications are working on your phone. They can download your survey answers, with your email address, as a spreadsheet for analysis; a spreadsheet they have downloaded is held by your research team under your consent form, outside Spark. Console does not yet show your location, Apple Health or surroundings data; this policy will say so when it does. Researchers see only the studies they have been given access to, and cannot see participants in other studies.
Us. As the operator of Spark, we can access data across studies to run, support and secure the service, and to carry out the requests in section 9.
Service providers. We rely on a small number of providers who process data on our instructions and are not permitted to use it for their own purposes:
- Amazon Web Services — hosting, database storage, account sign-in, the sign-up and password-reset emails Amazon Cognito sends you, and the network protection that screens requests to our servers.
- Expo — gives your phone the address it receives notifications at, and relays notifications to Apple and Google. Each notification’s title and text, your random account identifier, the identifier of the prompt and the date it belongs to, and your phone’s notification address pass through Expo on the way to your phone.
- Apple Push Notification service and Google Firebase Cloud Messaging — deliver notifications to your device. The same notification content passes through them.
- Apple and Google maps — when Spark shows a map, your phone loads it from Apple Maps (iOS) or Google Maps (Android). When you place a work or school pin, your phone asks Apple or Google for the address at that spot. These requests carry the coordinates being shown or looked up.
Apple Health data is not shared onward. Health and activity readings are used for your study and are never disclosed to advertisers or used for marketing.
Legal disclosure. We may disclose data if required by law, or where disclosure is necessary to prevent a serious and imminent risk of harm. If your consent form describes a specific reporting obligation, that obligation applies.
Research outputs. Findings are published in aggregate. Individual participants are not identified in publications.
7. Where it is stored and how it is protected
Study data is stored on Amazon Web Services infrastructure in the United States (region us-east-1). Requests from your phone may enter AWS’s network at a location nearer to you on their way there. If you are outside the United States, your data is transferred to the United States and is subject to United States law.
- All traffic between the app and our servers is encrypted with HTTPS.
- Data on our servers is encrypted at rest.
- Reading your data requires your own signed-in account, a researcher’s account limited to the studies they have access to, or our own service key.
- Study records are stored under a random account identifier rather than your name or email address.
No system is perfectly secure. If a breach affects your data, we will notify you and the relevant authorities as required by law.
This website uses no analytics, advertising or tracking scripts and sets no cookies. It remembers your light or dark theme choice in your own browser, and the design page keeps a sketch you have not sent in that browser tab. Our hosting provider, Amazon Web Services, records standard request details such as IP addresses when you visit.
8. How long it is kept
Research data is kept until it is deleted at the end of the retention period stated in your study’s consent form and required by the ethics board that approved it, or until you ask us to delete it (section 9). Spark does not delete research data on a timer. Research retention periods are typically measured in years, because published findings must remain verifiable. An account that never joins a study is kept until you ask us to delete it.
Some records expire on their own:
- Server logs: 90 days.
- Records of background tasks that failed, kept to diagnose the fault: 14 days.
- Database backups, which let us recover from a fault: 35 days.
- The ticket we hold for each notification while we ask Expo whether Apple or Google accepted it: removed as soon as Expo answers, usually within the hour, and set to expire after 3 days if it never does (the database can take a few more days to remove an expired ticket). The answer itself, whether Apple’s or Google’s notification service accepted the notification, is kept with your prompt records as research data.
If we delete your account at your request, your sign-in account and every record stored under your account identifier are removed, as described in section 9. Copies in backups, server logs and failed-task records then expire on the schedule above. Our log of staff actions on your account is kept, and spreadsheets your research team has already downloaded are held by them (sections 6 and 9). Data already de-identified and incorporated into a completed analysis cannot be traced back to you and therefore cannot be withdrawn from it.
9. Your choices and your rights
- Turn off collection
- Revoke any permission in your device settings at any time, and Spark stops collecting what that permission gives it. With notifications turned off, your phone still reports that they are off, so your research team can tell why prompts are not arriving. In the app you can adjust your prompt hours and, if your study allows it, pause prompts for up to an hour. Location tracking stops when you stop it on the screen you started it from, when you withdraw Spark’s background location permission, or when you sign out.
- Get a copy of your data
- Email us at the address below and we will send you a machine-readable (JSON) copy of the records we hold about you: every study record stored under your account identifier, your account details, and our log of staff actions on your account.
- Delete your account
- Email us to delete your account. We remove your sign-in account and every record stored under your account identifier: enrollment, responses, ratings, locations, surroundings, health and activity, push tokens, notification and delivery history, scheduling records, and settings. It cannot be undone. Some records outlast it. Our internal log of actions staff took on your account (such as exporting your records, or sending you a prompt by hand and whether it was sent) is kept indefinitely; it identifies you only by your random account identifier, but also records your study, when staff acted, and the staff member. Spreadsheets your research team has already downloaded are held by them under your consent form (section 6). Copies in backups, server logs and failed-task records remain until they expire (section 8).
- Correct your data
- Ask us, or your research team, to correct anything inaccurate.
- Withdraw from the study
- You may withdraw at any time, without giving a reason and without penalty. Contact your research team, whose details are on your consent form.
- Complain
- You can raise a concern with us, with the research ethics board named on your consent form, or with your local data protection authority.
Depending on where you live, you may have further rights under laws such as the GDPR, PIPEDA, or the CCPA, including the right to restrict or object to processing. We honour these requests regardless of where you live. We respond within 30 days and do not charge for it.
10. Children
Spark is intended for adults taking part in research and is not directed at children. It is not a consumer app: it does nothing useful without a study ID issued by a research team, and it has no content or purpose aimed at children. We do not knowingly collect personal data from anyone under 13. If you believe a child has created an account, contact us and we will delete it.
Eligibility to join a study is set separately, by that study’s protocol and its ethics approval. Where a study enrolls participants under the age of majority, it does so under that approval and with the consent of a parent or guardian.
11. Changes to this policy
If this policy changes, we update the date at the top of the page. If a change materially affects what we collect or how we use it, we will notify enrolled participants in the app or by email before it takes effect. Material changes to the research itself additionally require ethics board approval and, where applicable, your renewed consent.
12. Contact
Spark is published and operated by Harvey Wang, trading as Entropic Health. For privacy questions, data exports, or deletion requests: harvey@dal.ca
For questions about the study you are enrolled in, including withdrawal, contact the research team named on your consent form. For concerns about how the study is being conducted, contact the research ethics board named there.