6 ms·
Decades-old IBM database became profitable dossier on health of 270M Americans
- hedora 5y ago> He started by reaching out to the biggest corporations. If they would agree to give him data on their employees’ paid medical claims, he would return to them an analysis of their cost drivers, benefit designs, and manageable risks that would give them leverage in negotiations with insurers What? Isn't this exactly the sort of thing HIPAA is supposed to ban? What happened to doctor-patient confidentiality? Why do employers even have that information?
- capableweb 5y agoThe clues are a bit earlier in the article and the full name of HIPAA > The financial trajectory of MarketScan was perhaps unimaginable in 1981, when a former insurance executive named Ernie Ludy founded the company. His idea was to simply collect patients’ data and parcel it out to big companies that were seeking to control costs by getting a more granular view of their employees’ health care use. > The Health Insurance Portability and Accountability Act of 1996
- hedora 5y agoAhh, thanks. Should have kept reading. However, a paragraph or so later, it says HIPAA doesn't apply to de-indetified data, and that it's easy for researchers to buy the data set. Hopefully, some security researchers will get their hands on it, de-anonymize the data set, and then regulators will burn the industry to the ground.
- nerdponx 5y agoOr it will just keep on going with the HIPAA fig leaf like before. See also: credit rating agencies, ad targeting.
- ghaff 5y ago>HIPAA doesn't apply to de-indetified data Although I'd point out that very little is needed to un-deidentify medical records if you want to. For example, see some of the work Latanya Sweeney has done. http://latanyasweeney.org/ http://latanyasweeney.org/ https://arstechnica.com/tech-policy/2009/09/your-secrets-live-online-in-databases-of-ruin/ https://arstechnica.com/tech-policy/2009/09/your-secrets-liv...
- kapuasuite 5y ago> Hopefully, some security researchers will get their hands on it, de-anonymize the data set, and then regulators will burn the industry to the ground. Who does this benefit, though?
- zitterbewegung 5y agoWow, this person had a Data Science company in 1981 that was actually profitable.
- nexuist 5y agoComputer science evolves in a circle. Analyzing data for meteorological predictions was one of the first uses of a digital computer; it just wasn't hip at the time and involved 0 pandas.
- rubatuga 5y ago> import tensorflow as tf
- brilee 5y agohttps://en.wikipedia.org/wiki/FICO https://en.wikipedia.org/wiki/FICO how about 1956?
- gravypod 5y agoIANAL but I think HIPAA: 1. Only applies to covered entities. 2. Data can be given to another covered entity as long as certain rules are followed. Very easy to have a web of people giving each other data because of this.
- NightMKoder 5y agoYep and covered entities are usually related to billing for medical care. As an example, almost all life insurance companies are not even hipaa compliant because they aren’t covered entities. In net - hipaa doesn’t protect medical information generally - only the subset that’s usually visible to doctors. And even then, it stops working as soon as the info is outside a covered entity.
- Spooky23 5y agoHIPPA is a joke and basically stopped gossip. “Hot” conditions with high conversions are tracked in near real-time. I learned this when we received via FedEx a box of Enfamil on what should have been the due date of of daughter. Unfortunately, we miscarried.
- LancerSykera 5y agoAs a fedex driver I always dreaded delivering those. Not because of the possibility of unfortunate outcomes such as yours (never crossed my mind), but because I knew they were unsolicited "gifts" that came from big data schemes like this. Just like how marketing companies know that you're pregnant even before you do.
- walterbell 5y agoPotential use case for guerrilla sticker educational program.
- wolverine876 5y agoAny insights? How often did you see them? How did you know the recipient didn't order them?
- loeg 5y agoHIPAA (one P, two As) is chiefly for data portability between doctors; not privacy. HIPAA is not intended to and does not protect your medical information from insurers. Sorry to hear about your daughter. That is really tough.
- thaumasiotes 5y ago> HIPAA (one P, two As) There's a weird thing going on where everyone pronounces it as if it were spelled "hippa", and then everyone believes it must be spelled that way because of how it's pronounced. It doesn't seem to have occurred to anyone to pronounce it in a way that's compatible with the spelling.
- icegreentea2 5y agoThe information is supposedly de-identified. That means it's not a HIPAA violation. You can look at the HIPAA de-identification standard right over here (https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de-identification/index.html#standard https://www.hhs.gov/hipaa/for-professionals/privacy/special-...).
- alexb_ 5y agoI will never believe someone when they say info is "de-identified", because even if it is, it is shockingly easy to pinpoint 1 person out of millions based on a very small number of unique factors.
- nerdponx 5y agoA lot of people simply don't know this. Politicians either don't know or claim to not know. This is a good example of the law and popular conception of a concept being badly out of date, to the advantage of industry and disadvantage of regular people. Therefore industry has a vested interest in keeping the public perception focused on "de-identified" with a narrow definition of PII.
- tomrod 5y agoThere are academic subfields that study the risk of de-identification.
- samhw 5y agoI think it’s because the word “anonymous” permeates our understanding of the concept, and – even for people who didn’t actually study Greek – the tie to “name” is clear. We need to understand that “knowing someone’s identity” is not coextensive with “knowing their name”, and that in fact knowing someone has a rare medical condition may be more identifying[0] than knowing their name. [0] Or k-deanonymising, for anyone who’s pedantic about identity being an absolute.
- walnutclosefarm 5y agoI spent a good deal of my professional life in the last decade dealing with the problem of de-identifying medical data. You are correct that it's hard, but the HIPAA rule is actually not a bad go at it. See https://www.hhs.gov/hipaa/for-professionals/privacy/special-topics/de-identification/index.html https://www.hhs.gov/hipaa/for-professionals/privacy/special-.... For the kind of data in this particular database (mostly insurance claims data), it's highly unlikely that you could learn much through re-identification of the data.
- Froedlich 5y ago> What happened to doctor-patient confidentiality? It was cast by the wayside long ago. First it was the "mental health" exemptions, then various law enforcement provisions, then third-party "office solutions", then transcriptionist services, then private medical databases... You might as well assume that anything you tell your doctor is going to be recorded and (eventually) used against you. And then, as was alluded to earlier, there's the problem of incorrect information in those databases... and your only recourse is to give them more-accurate information to sell. All the dice are loaded against the patient. And as anyone online for long knows, data is forever.
- walnutclosefarm 5y agoThe data is de-identified, and thus not subject to HIPAA restrictions. It's not made entirely clear in the article, but most of this data is insurance claims data, not medical records per se. That's why employers have it. If your employer underwrites your medical claims directly - which most do nowadays - when you or your doctor submits an insurance claim, they are submitting it to your employer. It may go through a health insurance company - since most employers hire one to administer their plans - but that insurance company is collecting the information on behalf of the plan owned by your employer. The fact that it's insurance claims and not raw medical records is one of the challenges IBM had in making a business out of analyzing it. There is a lot less and less quality, medical data in insurance claims than IBM hoped.
- peterth3 5y ago> Medical data mining companies have made a business of scraping the details of consumers’ daily lives into medical dossiers that, if combined with MarketScan’s de-identified information, could be used to re-identify the individuals within its databases. De-identification is unreliable. If you have enough context, then the patients can be re-identified.
- walnutclosefarm 5y agoI agree that de-identification is imperfect, and re-identification is a risk (re-identification is, however, illegal in several states). However, the notion that there is big business in re-identifying medical records to target individuals is largely a myth, because there isn't a great market for individual-level information. Life insurance companies can't use it - they are required to get information they use for underwriting from the applicant, or with the applicant's permission. Companies who market medical products generally won't or can't use it (for one thing, it's typically not very timely - by the time your diagnosis appears in a re-identified data set, the time to get your attention is pretty well past for most conditions). The companies that came seeking medical records (I was until recently Chief Technology Officer for a clinic/hospital system with 10M active patient records) were satisfied with de-identified data, because they were looking for insights to develop new, or re-purpose existing, therapies and technologies, or to create marketing strategies for their tech. They had no need or desire to know who a particular patient was, because they were looking for broader insights.
- sha256sum 5y agoThere’s also MIB Group (Medical Information Bureau) which collects healthcare data as part of an exemption to the fair credit reporting act, along with an extensive astroturfing campaign to hide their activities (the Wiki article on them is useless). > In addition to an individual's credit history, data collected by MIB may include medical conditions, driving records, criminal activity, and participation in hazardous sports, among other facts. MIB's member companies account for 99 percent of the individual life insurance policies and 80 percent of all health and disability policies issued in the United States and Canada. https://www.ftc.gov/news-events/press-releases/1995/06/medical-information-bureau https://www.ftc.gov/news-events/press-releases/1995/06/medic... You may request the data they have on you and allegedly you can dispute the information.
- SilasX 5y agoTangent, but ... they picked the acronym MIB? I had to look it up, and the term Men in Black in the lay usage dates to 1947.[1] But the group MIB dates to 1902, so just a coincidence. [1] https://www.history.com/news/men-in-black-real-origins https://www.history.com/news/men-in-black-real-origins
- rr808 5y agoJust imagine the FB database in 20 years. Maybe its open by then
- Tempest1981 5y agoI wonder what they plan to do with our data? I found this: - Francisco Partners is a leading global investment firm that specializes in partnering with technology and technology-enabled businesses. - FP's current and past investments include such companies as BeyondTrust, ClickSoftware, GoodRx, Ichor Systems, iconectiv, LegalZoom, Quest and Verifone. and from the article: - Francisco Partners had previously purchased stakes in the telemedicine and drug coupon company GoodRx, the virtual appointment booking company ZocDoc, and Edifecs, a company that builds software to enable a more seamless exchange of data. - The firm declined to comment on the acquisition or its plans for the MarketScan database.
- walterbell 5y agoQuite a portfolio: - BeyondTrust: admin credentials - ClickSoftware: service chain mgmt - GoodRx: prescriptions - LegalZoom: LLC identity, IP - Quest: security/identity, Erwin data modeler - SonicWall: firewall - Verifone: retail POS transactions - ZocDoc: doc-patient calendar
- airstrike 5y agoThere are literally dozens of PE firms and VC funds investing in Healthcare IT... and not necessarily for nefarious ulterior motives
- walterbell 5y agoUK Health Security Agency (HSA) has a comprehensive page on test & trace data management, https://www.gov.uk/government/publications/nhs-test-and-trace-privacy-information/test-and-trace-overarching-privacy-notice https://www.gov.uk/government/publications/nhs-test-and-trac... (Dec 2021 update) > Under data protection law, you have several rights over your personal information. You can exercise any of your rights by contacting us at InformationRights@UKHSA.gov.uk > COVID-19 data may need to be shared with WHO for research purposes and where required to help trace contacts internationally. These are restricted transfers made on the basis of being important for reasons of public interest, where we rely on one of the derogations under Article 49(1)(d) of the GDPR. > Personal information includes ... name, DOB, address, employer, locations visited, travel itinerary, mental health, lifestyle, social circumstances, ethnic origin, DNA/biometric data ... We send personal information to ... Amazon, AWS, Deloitte, MoD, NCSC, ONS, Police, Palantir, Serco, WHO [and others]
- smrtinsert 5y ago> IBM’s efforts to use the repository to transform broad swaths of the health care system ultimately fizzled. The company struggled to create the cloud storage and computing infrastructure needed to combine all the data so it could be analyzed by its AI and analytics machinery. Wow still? I'm a little surprised by this as almost as lots of cloud offerings these days seem designed for massive scale.
- wolverine876 5y agoConsider what is collected today: If you video chat with doctors, look at the privacy policies of the video chat platform. IME, it doesn't differ from most other applications: We collect what we want, including medical data; we share it with partners for marketing, etc. And of course the only clause that matters: We can change our policy at any time by posting a new one to this page (i.e., without notification, effectively), so effectively we have no policy and can do whatever we want. Someone I know needed a prescription medication that they didn't want in public records, for understandable reasons (I would be concerned too). It wasn't a 'controlled substance' like a narcotic; what they wanted was perfectly legal and appropriate; it was just a prescription drug that implied a certain underlying condition and related events, and they wanted privacy. They asked me for suggestions. I suggested that the best they could do was to find a small, independent pharmacy; call ahead to verify that the pharmacy didn't share its data; and pay with cash (apparently ID was required to obtain any prescription drug). They did all that and reconfirmed all that at the pharmacy. The next time they were at a doctor in the giant local health co, the nurse asked about the drug, which appeared in the health system's medical records. Later, for unrelated reasons, they obtained their own credit report and it included the drug (!). Epilogue: When I go to the doctor these days, as I'm signing in they tell me to sign on a small touch-sensitive pad. The pad is about the size of a signature line and displays nothing. What am I signing? That I received their privacy notifications. (I didn't omit a step - they never mentioned or offered them otherwise.) My pharmacy is similar - 'sign', they tell me.
- lotsofpulp 5y agoI would need hard evidence to believe a credit report had someone’s prescription medication listed on it. Or anything non borrowing money related / identity / location history listed. https://www.usa.gov/credit-reports https://www.usa.gov/credit-reports Even the Medical Information Bureau report should not list medications: > Doctors, hospitals, pharmacies, and other health professionals can’t submit information to MIB. The report won’t include every diagnosis, blood test, or a list of your medicines.
- wolverine876 5y agoI actually saw it. It was in a separate section of the report. Also, how did their hospital chain obtain the data? Connecting dots - a dangerous form of reasoning - I guessed it was via this report. Looking around: https://www.consumerfinance.gov/ask-cfpb/how-can-i-find-out-whats-in-my-medical-payment-history-en-1837/ https://www.consumerfinance.gov/ask-cfpb/how-can-i-find-out-... How can I find out what’s in my medical payment history? You should contact the consumer reporting agencies that specialize in medical records or payments. These agencies may supply reports on your prescription drug purchase histories, medical conditions, data from your insurance applications, and data from other sources. Life insurance companies, for example, commonly use these reports to evaluate policy applications from potential customers.
- huitzitziltzin 5y agoDon’t overreact to this. It’s really not a big deal. I have worked with data of this kind for years. First of all, there are a bunch of datasets like this or very similar to it both public and private. Example: have you ever been to a hospital in California? There is a state agency called OSHPD which makes inpatient discharge records available to academic researchers (like me) and to others like insurers. Texas, Oklahoma, New York (iirc) and many many other states make similar data sources available. What’s actually in the data? Generally all of the ICD-9 or -10 codes generated by a visit. There may be very limited demographic information - race, ethnicity, zip code, age. That’s about it. (That’s what de-identified means here.) Sometimes it will tell you the source of admission (eg “arrived by ambulance”), and generally will tell you the source of the payment (eg “patient has private insurance”). Some of these datasets will let you track patients over time (TX has a version like that) within a year. It is extremely extremely difficult to match this information to actual people, assuming anyone even cared to do so (which almost certainly no one does!). For one thing, most zip codes are large and contain many people (generally only three digits of the zip are reported if there are very few records from that zip). More importantly, you just don’t have that much information about anyone in the data - you would have a row which essentially says: “Hispanic male age 50-54 from zip code 78201 who was billed for a knee replacement.” Good luck figuring out who that is! Even for very rare illnesses it doesn’t work… if you already knew someone in 78201 had X Rare Disease and went to Y hospital, you might be able to find the record in the data with some work. But what additional information would you get? Maybe how much his insurance was billed?? You already had to know the guy had a rare disease just to find him! Also - observe the fact that this dataset keeps getting sold. That tells you something about how useful it actually is in practice. I get that some MBA at IBM thought it would be great for “generating insights” about cost drivers and saving money. It’s really not. If you look for the “cost drivers” in this data you will see that long visits and complicated procedures are costly. And then your next step is…??? This data does not give you anything which will help you predict or avert such episodes. Neither IBM nor anyone else has actually come remotely close to solving this problem, right? I am not aware of a single effort which has even moved the needle on spending by 1%! After holding on to it for a few years, IBM or whoever owns it next will talk up its value and sell it to another sucker. (If it were really worth that much to IBM, they wouldn’t sell it!) If you want proof that there isn’t a privacy concern here, get exactly this data from Texas (which makes it publicly available). Open it up and let me know how easy a time you have finding anyone you can identify. Even if you have friends and family in the state who you know went to the hospital you’re going to have an extremely hard time.