group stringclasses 5
values | figure stringlengths 3 12 | label stringlengths 15 72 | note stringclasses 5
values | source_id stringclasses 5
values | source stringclasses 5
values | source_url stringclasses 5
values |
|---|---|---|---|---|---|---|
rise | 1.42 → 20.3 | new cases per million person-years, before and after March 2020 | An interrupted time series across 65 million patient records. The cut-off is the start of the pandemic, not a COVID diagnosis. | rise | Dulal et al., 2025 | https://pubmed.ncbi.nlm.nih.gov/39775762/ |
rise | 4.21 → 22.66 | new POTS cases per month in the same records | Roughly five times as many people receiving the diagnosis each month. | rise | Dulal et al., 2025 | https://pubmed.ncbi.nlm.nih.gov/39775762/ |
rise | 31% | of severely affected long COVID patients were diagnosed with POTS | Of 467 people still on at least half-time sick leave a year after infection. Another 27% had orthostatic symptoms without meeting the criteria. | longcovid | Björnson et al., 2025 | https://doi.org/10.1161/CIRCEP.124.013629 |
scale | 1–3 million | Americans estimated to have POTS before the pandemic | null | di-pots | Dysautonomia International | https://www.dysautonomiainternational.org/page.php?ID=30 |
scale | ~80% | are women, most commonly between 15 and 50 | null | di-pots | Dysautonomia International | https://www.dysautonomiainternational.org/page.php?ID=30 |
scale | 94% | female in the largest patient survey, with a most-common onset age of 14 | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
diagnostic_delay | 2 years | median wait from first symptoms to diagnosis | Half wait longer. A quarter wait more than six years — the distribution has a long tail, which is why the "average" quoted elsewhere is so much higher. | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
diagnostic_delay | 25% | are diagnosed in the first year | In the same survey the average wait was 5 years and 11 months. | di-delay | Dysautonomia International, 2013 patient survey | https://www.dysautonomiainternational.org/page.php?ID=184 |
diagnostic_delay | 27% | saw more than ten doctors before someone named it | null | di-delay | Dysautonomia International, 2013 patient survey | https://www.dysautonomiainternational.org/page.php?ID=184 |
diagnostic_delay | 50% | travelled more than 100 miles to get POTS care | null | di-delay | Dysautonomia International, 2013 patient survey | https://www.dysautonomiainternational.org/page.php?ID=184 |
told_instead | 69% | were given an anxiety diagnosis before POTS | null | di-delay | Dysautonomia International, 2013 patient survey | https://www.dysautonomiainternational.org/page.php?ID=184 |
told_instead | 59% | were told their symptoms were "all in your head" | null | di-delay | Dysautonomia International, 2013 patient survey | https://www.dysautonomiainternational.org/page.php?ID=184 |
top_symptoms | 99% | report lightheadedness | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
top_symptoms | 97% | report racing heart | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
top_symptoms | 94% | report nearly fainting | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
top_symptoms | 94% | report headache | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
top_symptoms | 94% | report trouble concentrating | null | shaw | Shaw et al., 2019 | https://doi.org/10.1111/joim.12895 |
POTS open data: where to find help, and what the evidence says
Two small, hand-checked datasets about POTS (postural orthostatic tachycardia syndrome), published by the team behind SPLYT: POTS, a free iPhone and Apple Watch app for people living with POTS.
Free to reuse under CC BY 4.0. Please credit "SPLYT: POTS (splyt.fit/pots)".
What's here
| File | Rows | What it is |
|---|---|---|
data/pots-help-directory.csv / .json |
30 | Patient- and clinician-run organisations that help people find a POTS-literate doctor, get diagnosed, find community, get long-COVID care, or support someone. Covers 10 regions. Every link was opened and read by a person on the date in checkedOn. No individual doctors are listed. |
data/pots-figures.csv |
17 | Headline numbers about POTS: how fast diagnoses rose after 2020, how long diagnosis takes, how often people are told it's anxiety, the most common symptoms. Each row links to its source. |
data/pots-treatment-evidence.csv |
9 | Non-drug measures and off-label drugs, each with the evidence grade the guideline or trial gives it. |
data/pots-sources.csv |
11 | Every study and guideline cited, with sample size and a link. |
data/pots-figures.json |
— | All of the above plus the diagnostic criteria, in one file. |
House rules
- Every figure traces to a study, guideline or patient-organisation survey in
pots-sources.csv, with its sample size. No figure is rounded up or taken from a secondary blog. - Where the commonly quoted number is more flattering than the paper, we use the paper's version. The diagnostic delay is a median of 2 years with a long tail (a quarter wait 6+ years), not "an average of 6 years". The exercise-registry result is 71% of the people who finished, out of 251 who started.
- This is not medical advice. It cannot diagnose anything. Treatment rows report what a named guideline or trial says, never an instruction.
Read it on the web
- Find help: https://splyt.fit/pots/help
- The numbers, with the papers: https://splyt.fit/pots/data
- Where to start, including the 10-minute stand test: https://splyt.fit/pots/start
Source repository: https://github.com/abrarhussain-00/pots-open-data
Rebuilding
The data is generated from the source files behind splyt.fit/pots:
node --experimental-strip-types scripts/build.mjs <path-to-splyt-web>
About SPLYT: POTS
SPLYT: POTS helps you track POTS against your own normal. It runs the stand test on your watch over three mornings, tracks fluids and salt against a target, logs how you feel in one tap, and turns it into a two-page report for your doctor. One "I need you" button reaches the people you choose. It's free, with no ads. It does not diagnose or treat any condition.
Citation
See CITATION.cff. Corrections are welcome: open an issue.
- Downloads last month
- 55