Reasons for and against self tracking and quantification

Topic Infobox Question-icon.png
Linked pages on this wiki Tools (0),

Projects (0),

People (0)

Though fitness bands like Fitbit are quite popular, health tracking has a lot more to offer. This page is mainly intended to encourage more tracking. It is based off an argument map.[1] If you are not convinced by argument, please read Self Tracking Movements for a show of popularity of the practice or Interesting Results for examples of successes. Sections of this page are ordered by similarity, not importance.

Reasons ForEdit

Contribute to scienceEdit

Self-experimentation is still science. Variety of literature shows the validity of singe subject and patient-led studies.[2] 12 self-experimenters won Nobel prizes.[3]

Alternatively, you could donate your data. Papers often encourage greater granularity in the data they use. Even if no large formal experiment is conducted, observational studies are important.[4] Epidemiological studies[5], which are observational studies, become better with self-tracking tools.[6][7] Continuous self-tracking data from multiple people qualifies as longitudinal study. These are more reliable than single slice observational studies like the ones most often used by papers about diets. Unfortunately, mobile health companies that aggregate data from users rarely do it with real scientific goals in mind. Instead, you should give that data to researchers through Open Humans.

Manage health conditionsEdit

Doctors will get much more precise, objective, and detailed description from data than patients provide verbally. Patient knows much more about themselves than doctors have time to understand without the clear information that self-tracking provides. Data like this has been good enough to prove something even to the medical community. Patients get a clearer picture of their health which they can use to make better decisions.[8] Patients can gauge their doctor's performance better. Self-tracking could reduce medical costs.[9]

Sometimes doctors cannot help and then DIY health is the only answer.[10] Chronic or subclinical health conditions, and effects of lifestyle choices, like veganism,[11][12] may be monitored using personal health tracking. Remedies such as weight loss diets and melatonin can be tested for effectiveness with health tracking. Even if no remedies are found, tracking can provide piece of mind by showing the patient that this condition is like previous ones that have resolved or that it is benign.

Unfortunately, issues like this are often esoteric and so require elusive or underdeveloped tools. At the same time, serious chronic conditions are more likely to be tracked with expensive, restricted to the public, or closed data devices.

Protect your healthEdit

Even if you have no health problems and do not intend to improve anything, you should continuously monitor the vitals (same as in 'improve' bellow) that your doctors do not. Detect problems before they do harm. Sickness can be subtle.[13] It can harm you in a way that you do not notice but that could affect you a lot. For example, slight dietary changes could increase resting heart rate or more noise could drastically reduce productivity.

Unhealthy products are ubiquitous. Many foods are designed to be sold, often without concern for the health of the customer. Technological innovation grows so fast that culture and regulators (FDA) cannot keep up.[14] This turns in to a Digital addiction epidemic because of games like flappy bird, and apps like Facebook and Youtube.

Travel can be unhealthy in unexpected ways. Quality of sanitation and health care varies between countries. Health advisories about foreign countries are abundant. Digital nomads, or anyone that travels great a lot, should maintain a diary of symptoms to notice uncommon diseases. Productivity can suffer unnoticeably, because of stress or new environment affects traveler's judgement.

Long term monitoring provides safety baselines for self-experimentation. Just imagine someone trying out long term Fasting or nootropics and not noticing dangerous changes to their normal state. In addition, if someone who does not experiment ever decide to experiment, they will have lots of baseline data to compare against data gained during interventions.

Tools to save your life usually also record data. For example, Apple watch monitors for arrhythmia, medicine trackers for the goal of safety can also help monitor health conditions, and hiking is made much safer with GPS.

Improve Fitness, Mood, Productivity, and CognitionEdit

For example: What is the optimum amount of coffee to improve productivity but not affect sleep? This is very similar to trying to improve a chronic health condition, except that it can be useful to anybody and does not deal with critical issues.

Unfortunately, many industries that promise such improvements are filled with conflicting information and scams.[15] Consumers avoid buying products from such industries because they are not sure if there was any benefit from any specific product. Tracking of outcomes fixes this problem and encourages consumers to search for better products.

Humans adapt and change so constant self improvement needs constantly updated goals. Goals are made based on user;s history.

Health tracking encourages experimentation and constant self improvement. People have gone through their entire life without finding drastic life improvements because they simply did not experiment systematically. What if you are that kind of person?

This method has precedents outside of personal science. Many professionals, especially coaches of athletes, measure and try to improve every little thing to boost overall performance.

Motivate and change your behaviorEdit

When someone has successfully figured out what needs to change, they then must build up motivation to enact the change.This is the focus of the Habits and self improvement community. Tracking is useful here too. Do not dismiss failure as a lack of willpower because willpower is still brain chemistry and easily manipulable by outside factors. For example; sometimes lack of willpower is caused by health problems, sometimes just a bit too much friction can shut down motivation and productivity, or making a habit mildly more fun makes the habit something to look forward to. Avoid relying on willpower and instead get yourself to want (motivate) the new behavior.

Health tracking, (or rather mood tracking), can detect problems and determine the best behavior change technique for each individual.

Tracking is itself motivating. Most smart watches, the most popular wearable, produce lots of motivating information and nudges. Share accomplishments with friends to get external motivation. Gamification, the idea of dressing up real world as a game to make real world things more fun, seems popular and it requires some tracking. The self tracker has access to many other behavior change techniques, including; cool new tech, cool art, competition, empowering healthful communities, financial punishment, and guilt. Just watching the numbers improve or even be recorded can be motivating and build healthy habits; according to personal experience DG (talk) and Atomic Habits book. Feeling in control, and mastering yourself and skills, can be strong intrinsic motivation.

Feeling good is not enough. Improvement should include hard data.

Unfortunately, when things are not going well, people will become demotivated and even hide the event from their history. The negative effects of self tracking on motivation is mentioned in section arguments against tracking.

Make a story from your lifeEdit

Storing important memories is the primary concern of Lifelogging movement. With the help of self tracking you can give your loved ones a better picture of yourself. Better stored memories improve reminiscing for fun, introspecting for wisdom, and reviewing your decisions for good planning. Even as pictures, Lifelogging is useful to the other goals here because it provides context for data.

For FunEdit

Enjoy playing with cool tech gadgets.[16][17] Learn a new skill. Satisfy curiosity by looking at the data.[18] Make art out of it and post to r/DataIsBeautiful. Make serious art out of it to bring attention to an issue.

Reasons Against and ProblemsEdit

Self tracking takes too much time and effort every day.Edit

Active Tracking is time consuming, though automatic data recording is usually not. Some things can only be tracked manually.

Studies and experience show people dropping even almost completely automated fitness tracking smart watches. If self tracking becomes a habit it will require less effort. Effort and motivation is hard to define, e.x. some people like Journaling and others do not.

If appropriate effort is not put in, variables will end up with too much noise and any statistical tests will require too much data. Self assessment is vulnerable to mood, bias, and ratings drift so it requires extra thought and attention. Users may forget to fill out entries.

Finding and setting up data sources takes time.Edit

User could follow someone else's made trail, such as reading this wiki, instead of researching everything themselves. Uncommon problems, and the exploration inherent in science, require custom data sources that have not been documented here. Learning enough about health and statistics takes time too.

Greed motivates many companies to make fake and useless devices and interventions.[19] Many of the issues described on this page, data quality (devices often cover up bad signal with guesswork), data oversimplification, data access[20], and price are all constraints which make finding a good enough device harder. Medical and Open Source devices have their own issues. Afterwards the explorer will have to aggregate data (Open Humans), parse new data format and clean the data.

Recording is not necessary to find relationshipsEdit

Why take the time to write things down when user could notice relationships just by paying attention? Its quicker, easier, and far fewer of the reasons not to self track would apply. However, many of the reasons for self tracking can not be satisfied without recording data and most of the relationships will not be noticed as well. Most relationships user would want to uncover require statistical analysis, especially strong effects that can only be seen across long periods. Though, more important effects are usually more sever and easier to notice. Many things a user would want to track require a device and can therefor be automatically tracked with very little effort on the user's part. Noticing relationships relies completely on Self assessment with all its issues and minus a few of the mitigating techniques. This alternative to self tracking is easier though rarely usable.

User could arrive at wrong conclusionsEdit

Wrong conclusions could be dangerous to user's health. User should of course double check with experts like doctors and veterans of the community (and this wiki). Learning enough about health, experimentation,[21] and statistics takes time that many do not have.

Analysis algorithms are either hard to use or too incapable. For example, article in Nature where ML was used in multiple N-of-1 studies and that approach is both incomplete and difficult for the average user.[22] All data aggregators for self tracking, besides OH, use linear regression or nothing at all. This problem can sometimes be avoided with careful experimental design like RCT.

Sources can be misleading. Food companies will give bad data to make themselves look good. Even user can have biases for various reasons listed in "negative engagement of user".

User's doctor may not be able to help. Use of self tracking is not wide spread through medical community. Except for a few key vitals, doctors do not use the types of data self trackers do. Doctors are not skilled in advanced statistics.

Users may focus not on the most important metrics but on those that are most measurable.

Unheathy psychological effects of trackingEdit

See also Psychological effects of tracking. [23]

User's mood may become dependent on success, eg sad when cannot run as fast as usual or do not get the chance to get their exercise.[24] This is the negative side of users becoming motivated by watching their stats grow. Motivation to do activities could become dependent on tracking. Users may become obsessed with data tracking and consequently take time from better activities. Seem easily solved, just add "obsession with self tracking" metric.

When someones' motivation is controlled from outside it overrides intrinsic motivation and causes resentment and apathy. However, many of the motivating techniques are only encouraging, involving, and engaging rather than domineering. Users should easily notice if they are successfully motivated or not, so only effective apps would be popular.

Unpleasant emotions stop trackingEdit

Insincerity of apps can frustrate users into quitting them.

Self tracking can feel like a tyrant; demanding and planning things for the user.[25] Self tracking tools tells user what their goals should be and that can be very irritating. Many apps do not personalize those goals, eg recommending loosing weight to pregnant people. Some push their own agenda, eg lastfm encouraging listening to more music. Apps can produce more psychological pressure to conform to perceived normal. Apps can contribute to legitimized bad ideals.

People do not want to find out somethings,[26] like a genetic predisposition to a disease they can not treat. They also do not want to be constantly reminded of their problems.

Health tracking presents privacy concernsEdit

Most consumers already give lots of data to companies, such as their location via GPS, social media, and their search and browsing history.[27][28][29][30] Employers already monitor employee health to optimize the performance of their employees.[31] It is possible to do basic cognitive and psychological assessment via background things like speed of typing and semantic analysis.[32] Also conversations can be summarized with deep learning and emotions detected[33][34] via cameras.[35] Companies regularly hoard customer data,[36][37] sell it to 3rd party aggregators, and get hacked.[38]

Never the less, self trackers can make more data available.[39] In particular, many devices measure stress and Mood through heart rate and Galvanic skin response, kind of like a polygraph.[40][41][42]

Such data has ended up in the hands of employers to squeeze the most out of their employees, as well as mortgage lenders, and insurers to adjust rates.[43] It was even used to support discrimination and as evidence in court.[44][45][46]

Self tracker has some options for defense. Data protection regulations grow.[47] Government classified Medical Devices make medical data and that data is under more protection than basic commercial device data. Data can be anonymized, though often its not hard to deanonymize it. User should choose companies with terms of service that require opting into user tracking programs. Open source apps, and some commercial ones, are under your complete control and will not mind if the device it runs on never connects to the internet. This wiki will note open source tools and tools with exceptional demand or lack there of for user data.

ReferencesEdit

  1. https://www.kialo.com/everyone-should-health-track---self-quantify-49787
  2. https://scholar.google.com/scholar?hl=en&q=patient-led+study
  3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3298919/
  4. https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis
  5. Introduction to Epidemiological Studies - PubMed (nih.gov)
  6. https://www.jmir.org/2022/6/e35804
  7. https://www.researchgate.net/publication/49775085_Reducing_case_ascertainment_costs_in_US_population_studies_of_Alzheimer's_disease_dementia_and_cognitive_impairment_-_Part_2
  8. https://dl.acm.org/doi/10.1145/1124772.1124910
  9. https://flowingdata.com/2022/06/24/when-americans-had-intercourse-with-opposite-sex-for-the-first-time/
  10. https://www.lesswrong.com/posts/fFY2HeC9i2Tx8FEnK/my-resentful-story-of-becoming-a-medical-miracle
  11. https://www.vegansociety.com/shop/veg-1-supplements
  12. https://en.wikipedia.org/wiki/Vegan_nutrition
  13. https://www.fda.gov/drugs/special-features/high-blood-pressure-understanding-silent-killer
  14. One-Third Of New Drugs Had Safety Problems After Approval : Shots - Health News : NPR
  15. https://disb.dc.gov/page/beware-weight-loss-scams
  16. https://www.asc-csa.gc.ca/eng/sciences/bio-monitor.asp
  17. https://spinoff.nasa.gov/Spinoff2019/hm_6.html
  18. https://blog.zenobase.com/post/107823078262
  19. https://www.jmir.org/2022/6/e37677
  20. https://quantifiedself.com/about/access/
  21. https://www.frontiersin.org/articles/10.3389/fdgth.2020.00003/full
  22. https://www.nature.com/articles/s41398-021-01445-0
  23. https://dl.acm.org/doi/10.1145/3274309
  24. www.wsj.com/health/wellness/sleeper-performance-people-obsess-over-their-stats-biometrics-6a6feff
  25. https://www.sciencedaily.com/releases/2016/09/160908141724.htm
  26. https://xkcd.com/2563/
  27. https://en.wikipedia.org/wiki/Big_data_ethics
  28. https://www.youtube.com/watch?v=w0NEEmQDVLI
  29. https://www.youtube.com/watch?v=fCUTX1jurJ4
  30. https://www.economist.com/the-world-if/2018/07/07/data-workers-of-the-world-unite
  31. https://publications.tno.nl/publication/34623617/ApIQVD/TNO-2015-R11632.pdf
  32. https://pubmed.ncbi.nlm.nih.gov/19963743/
  33. https://en.wikipedia.org/wiki/Emotion_recognition
  34. https://en.wikipedia.org/wiki/Emotion_recognition
  35. https://ieeexplore.ieee.org/document/8734245
  36. https://old.reddit.com/r/dataisbeautiful/comments/wkzqxs/oc_the_business_apps_that_collect_most_data/
  37. https://www.fiercehealthcare.com/health-tech/report-third-top-hospitals-websites-collecting-patient-data-facebook
  38. https://zerforschung-org.translate.goog/posts/datenabfluss-auf-rezept/?_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=sc
  39. https://www.mobihealthnews.com/news/most-health-apps-have-ability-collect-and-share-patient-data-study-finds
  40. https://en.wikipedia.org/wiki/Polygraph
  41. https://www.wareable.com/wearable-tech/wearables-that-track-emotion-7278
  42. https://www.heartmath.com/tech/
  43. https://www.tandfonline.com/doi/full/10.1080/03085147.2019.1570707
  44. https://www.forbes.com/sites/thomasbrewster/2018/01/29/strava-fitness-data-location-privacy-scare/
  45. https://mashable.com/article/period-apps-roe-v-wade
  46. flcourts.gov/content/download/881800/file/Pacemakers%20Fitbits%20and%204th%20Amendment.pdf
  47. https://en.wikipedia.org/wiki/General_Data_Protection_Regulation