Ask HN: Is there a webapp to input/store your blood test results?
Aside from excel/Google Sheets,
Does anyone know of a web app (preferably) that lets you manually input blood test results?
The hope is a service that creates graphs of historic records, provides reference values comparison, unit conversion, and offers insights (like what high/low values could indicate)?
60 comments
[ 2.7 ms ] story [ 117 ms ] thread1. People are far more likely to store and curate their vacation pictures than healthcare data. There is a behavioral aspect to this. There isn't a lot of 'feel good' attached to storing a good HDL/LDL reading. Quite the opposite when it's bad.
2. Most countries develop their own version of a EHR, sometimes controlled by patients. Success rates vary but they remain a very useful tool for managing chronic conditions and patient peri-operative journeys. Once governments (local or national level) take over, there is very little incentive for anyone to develop.
3. There have been many attempts to create a patient-led EHR by companies but it is difficult to get off the ground and sustain due to data ownership, security, sharing and commercial complexities.
4. Storing the data itself is trivial. FHIR covers this well and within haematology at least, the support for standardization is strong. LOINC, SNOMED and UCUM are well established. Just be careful to store the initial units and reference ranges.
5. My personal take is that personal EHRs will take off once a good open source p2p toolkit implements superb encryption-led access control across workflows with top institutions comes into play. This is unlikely because of how EMRs lock up provider data.
It may have been InsideTracker (https://www.insidetracker.com/) or Tally Health (https://tallyhealth.com/).
Some of the specific biomarkers you can input into the platform include LDL-C, HDL-C, Triglycerides, Non-HDL-C, VLDL-C, TG/HDL-C, Total Cholesterol, Lp(a), ApoB, ApoE, Homocysteine, hsCRP, Uric acid, Fasting Insulin, Fasting Glucose, OGTT, Hba1c, Campesterol, Sitosterol, Lathosterol, Desmosterol, TSH, fT4, fT3, rT3, ATA, ATPA, Estradiol, FSH, LH, SHBG, Testosterone, Free testosterone, PSA, ALT, AST, Total bilirubin, Cystatin C, eGFR - Cystatin C, Hb, Ferritin, Omega-3, and Vitamin D.
Once your blood test results are inputted, the system enables graphical representation of your historical records. This can help visualize trends over time, potentially revealing patterns or changes that might otherwise remain unnoticed. It also provides reference values for comparison, making it easier to understand where your results sit within the broader context (for things like gender/age).
It also tracks metrics across sleep, exercise, nutrition, emotional health using data from your wearables.
Not knocking it - as a non American I got a kick out of seeing imperial units being used out in the open :laugh:
EDIT: Oh, cannot sign up without a US address.
https://github.com/markwk/awesome-biomarkers#biomarker-track...
I've heard problems with people who use the tool but as a patient to be able go see a decade plus of my hospital visits, checkups, scans, and x rays is pretty amazing.
You are right that you cannot enter the data yourself, as far as I know. But if your lab or physician uses a system that can let you download the records, they will be stored. So I have access to all my blood labs, but there is no easy way to graph them.
As far as I know, there is no App Store rule against making an app that lets you log the variables manually, nor is there a rule against making an app that graphs stuff. Keep in mind that there are a lot of Apple rules around HealthKit data to protect privacy, and writing such an app could be difficult.
<sightly less cynical view> There's been persistent rumours for a few years now that Apple are working on blood glucose sensing for Apple Watch. If those rumours are right (and a big part of me really hopes they are) then that'd likely come with built in apps that destroy every blood glucose related 3rd party app.
Edit: I just noticed the original post wasn't specifically blood _glucose_ related. My own "blood test" requirements made me jump to an erroneous conclusion...
https://apps.apple.com/ca/app/heart-reports/id1448243870
No affiliation with the app, just a user.
[1]: https://picnichealth.com/
It's free if you have one the conditions they're looking for or they accept you to be part of a research group
However, I was surprised when I used it last. They were able to find doctors that I didn't report. Turns out a lot of doctors are part of networks in geographical areas that allow sharing of medical records. They might've also used my insurance provider. If you see doctors with insurance, the insurance company will also have a record.
I don't think doctors are required to report patient records to a central authority. The government is able to access medical records without a warrant though. They might have a centralized way of knowing which doctors a person has visited.
It's designed to automatically pull your medical records from your EMR -- but users can also enter their conditions, medications and procedures manually. We have plans to allow users to manually enter their lab results [2], however that feature has not been implemented yet.
Fasten is still a work-in-progress, but if you're willing to contribute we have a pretty active discord (linked in the README).
[1]: https://github.com/fastenhealth/fasten-onprem/ [2]: https://github.com/fastenhealth/fasten-onprem/issues/137
- Epic: https://open.epic.com/MyApps/Endpoints - Cerner: https://github.com/cerner/ignite-endpoints
It will be open sourced (eventually) -- once I've cleaned up some test credentials that were inadvertently committed to the repo.
Hope that answers your question?
However, we'll make sure to update the README to make it clearer that international providers are not supported yet.
We haven’t built historic import yet, but our members get to see all labs over time in a slick web interface.
What do you all think about the pricing and product from our homepage?
If you’re looking for another open-source and free alternative, I’m building Mere Medical which is not quite as developed as Fasten but takes a local-first approach and still connects to 2000+ healthcare institutions.
www.meremedical.co www.github.com/cfu288/mere-medical www.demo.meremedical.co
Mere is a PWA stores all of your records in localStorage in your browser, data is not stored on any server.
The goal is to be able to back up and restore encrypted blobs to external storage (s3, Dropbox, etc) to mitigate this and enable sync across devices, but this isn’t implemented yet.
Yes, it isn't a complete interoperable EMR solution, but I kinda manage it myself by downloading the PDF reports and maintaining in my local electronic filing system. There is always some esoteric test that the doctor could order which has to be done at a specialist lab. So, my low-tech solution settles on PDFs of the reports. It works for 99% of the cases.
At least the trend in India is to lock all of this data into proprietary data silos.
There is an initiative in India currently underway to provide a digital health ID and an ecosystem for healthcare participants to share data Ayushman Bharat Digital Mission (ABDM). It is yet to take off in a big way (and hopefully it does in the next 4-5 years) and achieves the same kind of success as the payments infrastructure by NPCI that brought the UPI network.
They do a lot of analysis. You can send them a PDF and it will (try to) OCR it and auto-fill which in my experience is 80-90% good so you still need to check and fill out a couple blanks but it mostly works.
[0] https://selfdecode.com/
We work with diagnostic laboratories in the Philippines. Besides blood tests, we let users input results for clinical chemistry tests, x-rays, and more. We're also able to pull results straight out from hematology analyzers and other diagnostic equipment.
We also provide delta checking (comparison of your current vs past results), reference range checks, and unit conversion. We don't offer insights, but it is something we can build with a large enough dataset.
Hit me up if you're interested!