[ Wednesday, May 27, 2015 ]
Beacon Health (South Bend, Indiana):
Jeff [10:09 AM]
Another day, another hack. Today's unlucky victim is Beacon Health.
It looks like only emails were compromised, and so far there's no actual evidence of misuse. No indication of how it happened, but I'd suspect phishing.
[ Friday, May 22, 2015 ]
Next Phase of OCR Audits:
Jeff [2:11 PM]
Have you received a survey notice from OCR in the last few days? It appears that the long-awaited second phase of audits if finally rolling out
. Nobody outside OCR knows for sure if these are all that are coming, or if everyone who gets a survey will get audited. So, if you got a survey, it doesn't necessarily mean you'll be audited (but it's more likely than if you didn't). And if you didn't get a survey it doesn't necessarily mean you're in the clear (but again it's more likely than if you got a survey).
[ Wednesday, May 20, 2015 ]
Latest Hack Victim: CareFirst BCBS.
Jeff [5:10 PM]
Doesn't sound so bad, but I've learned to wait for the rest of the shoes to drop.
[ Wednesday, May 13, 2015 ]
Indiana State Medical Association Breach:
Jeff [3:00 PM]
The IT chief of the ISMA, of all people, lost a laptop and two hard drives
containing Social Security Numbers and medical information of about 40,000 beneficiaries of the Association's health insurance plan. Apparently they were (i) unencrypted and (ii) left in plain sight in his unlocked car. In addition, he failed to report the theft for 24 hours.
Possible upside: ISMA's insurance is through Anthem, so maybe all of the data had already been stolen when Anthem got hacked.
[ Tuesday, May 05, 2015 ]
Is this a HIPAA breach? A guy had a leg amputated
Jeff [5:20 PM]
, and the hospital threw the leg in the trash, with the patient's name written on it. The cops found it in the landfill and, as you might expect, checked up on the guy, thinking foul play might've been involved.
One could argue that, while the name written on the leg is definitely an identifier, the leg itself is not "information" and therefore this could not be protected health information. However, the presence of the leg (with the idenfitier) implies
some information, even if it's not "information" itself.
I would suspect OCR would consider this to be PHI, based on past experience, but if you wanted to say it wasn't, I'd say you at least would have a leg to stand on.
Hat tip: Ron Holtsford
Baltimore Riots: Anyone know whether CVS suffered a data breach when their store was looted? An inquiring reader of the blog (from
Jeff [4:42 PM]
Birmingham Montgomery, AL) raised the issue, and it's definitely interesting.
My assumption would be that the store operates on some sort of dumb terminal pharmacy information system for the drug records, so that there's no real data stored in any of their drugstores; it appears on the in-store computers while they are being used, but isn't stored there, so that when the computers are powered down and disconnected from the central network, they don't have PHI. Of course, there would be some PHI in the form of paperwork, particularly in the bags of filled-but-not-purchased prescriptions. There might be some other paper records as well. And CVS should have disaster recovery systems to determine whose filled prescriptions were potentially taken, but I'm not sure how well they'd be able to tell if any other paper records were compromised.
Could be an interesting exercise at CVS right about now. . . .
Jeff [4:28 PM]
this time it's Partners Healthcare in Boston
, and 3300 patients are affected. The good news is that the EHR itself wasn't compromised, just some PHI in some email accounts (presumably internal emails only . . . ).
[ Tuesday, April 28, 2015 ]
Jeff [9:20 AM]
[ Monday, April 27, 2015 ]
Another Pharmacy Trashing Patient Information
Jeff [10:24 PM]
: This time it's a small compounding pharmacy (Cornell Prescription Pharmacy in Denver) rather than a national chain, but unshredded paper records in the trash
are the culprit. Importantly, the pharmacy did not have HIPAA policies and procedures in place. No known harm was done, but the fine was $125,000.
"Failure to implement any written policies and procedures" equals $125,000. Key word: any.
[ Monday, April 20, 2015 ]
Workplace Wellness Programs:
Jeff [12:57 PM]
Do you have one? Is it covered by HIPAA? Maybe, maybe not
Big HIPAA fines are coming
Jeff [11:11 AM]
. . . . I keep hearing this
, and I'm sure there will be some doozies.
[ Wednesday, April 15, 2015 ]
More on Medical Identity Theft
Jeff [11:55 PM]
: certainly getting a lot of attention
UPDATE: I've got to say that's a little misleading. "Nearly 60% were the result of theft" makes is sound like there were 600 breaches affecting about 20 million people, where the data was stolen for nefarious purposes. But that ain't true in the least. Yes, that many people had their data impacted by a theft, but the theft was not a theft of the data -- it was a theft of a laptop or flash drive or cell phone or some other piece of technology that could be sold (the equipment, not the data) for a profit. In virtually all of those cases, the thief had zero interest in the data. Those are "crackhead" cases, and in virtually all of those, the thief deleted or destroyed the data at the first opportunity. Theft, yes; theft of the data, not really.
[ Friday, April 10, 2015 ]
Another Individual HIPAA Criminal prosecution:
Jeff [1:49 PM]
this time a hospital respiratory therapist
who accessed patient data. No indication in the article what she was attempting to do with the data, but since it's a criminal complaint, I suspect either identity theft or personally-motivated snooping.
[ Tuesday, March 31, 2015 ]
Jeff [2:18 PM]
[ Monday, March 30, 2015 ]
Jeff [4:10 PM]
[ Thursday, March 26, 2015 ]
Medical Identity Theft:
Jeff [9:25 AM]
yet another story
on how it's the fastest growing type of identity theft. Some good points about your strategy for preventing it: it should include
- Encryption: at least consider it, but realize that sometimes -- for example, in the Anthem hack, the hackers got access to system administrator accounts, so they had the encryption key anyway.
- Data Loss Prevention: DLP encompasses several concepts, such as software to analyze data access and use, and systems to see when data is moved into or out of the system. It is always a good idea.
- Cyber insurance: check out prices and see if it's worthwhile to you. But make sure you know what you are buying: what is covered, what isn't, who pays first dollar, who picks the breach response and rehabilitation vendors, and where does the coverage end.
[ Tuesday, March 24, 2015 ]
A "Security Culture":
Jeff [4:25 PM]
does your hospital have one? Here are four traits
common to hospitals with a security culture, at least according to Sue Schade of FierceHealth IT. Although I think the middle two have a whiff of rent-seeking, I can't argue with 1 and 4.
[ Monday, March 23, 2015 ]
2015: Year of the Healthcare Data Breach.
Jeff [10:32 AM]
It's sure in the news
a lot right now.
[ Thursday, March 19, 2015 ]
What To Do If You're Hacked: These should be self-evident
Jeff [2:21 PM]
, but might not be. They are all elements in a decent "Breach Incident Response Plan." Do you have a BIRT? You should get one; email me and I'll give you some help.
Target: I hear (h/t Lynn Block) that Target has settled the class action for $10,000,000, offering up to $10,00 to any individual who can prove damages caused by the breach. That's certainly a lot cheaper than I would have expected, and I suspect it's well below the "brand damage" losses Target suffered already.
Jeff [1:44 PM]
[ Tuesday, March 17, 2015 ]
Jeff [6:07 PM]
Medical Identity Theft in the News: Here's a story
Jeff [4:30 PM]
on how healthcare data is a treasure trove for data thieves. But here's a much better one
(they got some really smart guy to talk about it). And I was also interviewed by Scott Crowder of KTRH Radio in Houston (AM 740) this morning, for inclusion on the radio sometime tomorrow.
[ Wednesday, March 11, 2015 ]
Jeff [2:57 PM]
Jeff [2:39 PM]
[ Wednesday, March 04, 2015 ]
Jeff [3:47 PM]
[ Wednesday, February 25, 2015 ]
Hippler (East Texas Hospital Data Thief) Gets 18 Months:
Jeff [8:08 AM]
Like Gibson years ago, a healthcare worker who stole PHI to use for fraud gets 18 months in jail
. Two interesting points: first, the US Attorney's office isn't saying what hospital he worked at, and in fact the case has been weirdly under the radar. Secondly, is there a better name for a HIPAA violator than Hippler?
[ Wednesday, February 18, 2015 ]
Health Data Identity Theft:
Jeff [10:55 PM]
Interesting article from NPR
on the black market for stolen health data.
[ Friday, February 13, 2015 ]
Jeff [2:24 PM]
Looking for de-identification tools? Here's a good place to start
, with some proprietary and open source options. Nice to see my hometown UT-Dallas in the mix here.
A big hat tip to Daniel Barth-Jones (@dbarthjones) for this link.
One More Anthem Thought: What if the hackers were really looking for a needle, not the haystack? What if they weren't after tens of millions of medical identities to conduct identity theft or something else, but were really looking for specific information on a handful of specific individuals, and only accessed the huge amount of data to cover their tracks? If it really was a hack by Chinese nationals operating under the guise of the Chinese government, wouldn't that make more sense?
Jeff [12:27 PM]
I'm not saying I believe it was the Chinese government, any more than I believe the Sony hack was the North Korean government. Or, if actually were, I'm not buying that the Norks were so torked about a Seth Rogan film they'd waste their resources hacking Sony. I'm betting if it was them, they were looking for something else, perhaps something they could use to extort someone.
I'm not paraniod. Really.
More Anthem: Lessons for IT Leaders.
Jeff [12:21 PM]
Encryption decisions (really, any protective decisions) can have much greater consequences tomorrow than you realize today. And it is very important that you know the value of your information: not just the value to you, but the value to a hacker. You may think the information you hold is mundane, but what you think doesn't really matter. What matters is what the predator thinks.
[ Thursday, February 12, 2015 ]
More on the Anthem Hack, and "What It Means":
Jeff [10:50 AM]
I've posted several posts
on the Anthem hack, and I'm not the only one. AHLA sent out an email to its HIT and payers, plans and managed care practice groups explaining the hack and the class-action lawsuits already filed. More news is out today, from experts who think 2015 will be "the year of the healthcare hack
." Maybe, maybe not, but the news does bring a few additional issues to mind:
First, as the AHLA email points out, the initial lawsuits and some of the initial reporting point to the lack of encryption as a big factor. Some have indicated that the Anthem hack may cause HHS to harden the encryption requirement of the Security Rule (as you know, encryption is not a required element, only an addressable one, and HIPAA covered entities are free to forego encryption if they reasonably determine it's not right for them). However, the hackers apparently got user credentials; even if the data had been encrypted, the hackers could have used the credentials to de-crypt the data. The fact that encryption would've been irrelevant probably won't stop those claiming encryption should become required, but it's worth considering.
Secondly, some of the reporting is highlighting the "monetization" issue, which I've always seen as the issue. The hackers probably don't want the data because they're going to use the data; they want it so they can sell it to someone else who will use it for identity theft. If that's the case, there is a multi-tier market, which could be good or bad: as the data changes hands, it's harder and harder to catch the initial culprit; on the other hand, if there are several steps between the point of theft and the point of use, there are several opportunities to put systems or safeguards in place to catch the actors and/or prevent the improper use. In other words, you might not be able to stop the thief, but if you can stop the purchaser from using the stolen data, the criminal enterprise falls apart. Something to consider.
Another issue I hadn't thought about previously: not only can the stolen medical identity be used to obtain needed healthcare services (an impostor uses the stolen identity to directly receive needed healthcare services), the stolen identity could also be used to obtain unnecessary services. I can think of two examples: a stolen identity could be used to obtain Oxycontin or other prescription drugs that could then be resold, or could be used to bill for services that are not actually provided. In both cases healthcare providers would be required to be part of the scam, either unwillingly (a convincing doctor-shopping patient gets painkiller prescriptions) or willingly (a doctor bills for services not provided), but that's not inconceivable. My previous thoughts focused on the receipt of actual, needed services, in which case the value proposition is harder to see (you need an ultimate purchaser of the stolen identity who currently needs healthcare services); however, that's not the case, since you could get prescription drugs to sell on the black market. I hadn't considered that.
Finally, I had recently heard that while social security or credit card numbers don't bring much more than a couple of dollars each on the black market anymore, a stolen medical identity might be worth $50. In today's news from Reuters, it seems that a stolen medical identity is now worth only $20. These aren't hard and fast numbers, but still, that's a pretty big devaluation. Maybe the supply of medical identities (and concommitantly, the amount of hacking) is growing so fast the price is dropping; maybe hacker buyers are determining that medical identities aren't all that valuable; or maybe there's really not that big a market of buyers out there after all. I have no idea, but it's worth considering.
[ Wednesday, February 11, 2015 ]
Jeff [6:40 PM]
[ Tuesday, February 10, 2015 ]
More Anthem Fallout: Is Healthcare Particularly Vulnerable to Hacking?
Jeff [11:12 AM]
There are a lot of people saying that; most of them stand to profit if you believe them (including me, in fact). The Anthem breach gives an opportunity for a bunch
of news articles
on just this point
. Let's consider this for a moment.
Much hacking and phishing is aimed at access to quick-value money: credit card numbers that can be used right away (with the victim perhaps not knowing about the use until the bill comes, or perhaps not even noticing it when the bill comes), actual bank account or financial acount data so current funds can be withdrawn, phony checks written, etc. In this type of hacking, the reward comes quickly to the hacker, but might be small change and is usually not a long-term proposition.
Some hacking is designed to allow for real identity theft: the hacker acquires a social security number and other information, impersonates the individual to obtain credit cards, car loans, even house loans, runs up big debts, and when the credit card company or bank tries to collect, the impostor is gone with the loot and the victim is left to try to prove that it wasn't him that got/used the credit card, loan, etc. The reward takes longer, but can be much bigger than snatching a credit card number.
With regard to both of these types of hacks, the victim, the bank or credit card company, and the vendor at which the stolen credit card is used are all incentivized to prevent the hack, since all of them stand to suffer substantial harm: the victim's credit might be ruined (or he might pay for something he didn't get), and the bank, the credit card company, or the later vendor might be left with the bill.
Health records sometimes contain credit card numbers, but often don't, making them not particularly useful for the first type of hack. On the other hand, health records usually contain social security numbers and other demographic data that can be useful for the second type of hack. Thus, medical records might be useful for traditional identity theft schemes.
The much bigger risk, and what medical records are particularly well suited for, is medical identity theft. This type of hack targets patients with good insurance, and allows someone to impersonate the insured and receive the insured's health benefits. The impostor gets free or reduced cost healthcare, but unlike most other hacks, the "victim" (the person whose data was stolen) doesn't necessarily suffer (or at least doesn't suffer immediately); in fact, the victim might benefit, since the impostor might actually pay a part of the victim's annual deductible. Additionally, the person whose data was stolen is not in a very good position to know it was stolen, unless he regularly checks his EOBs (frankly, even if he scrupulously checks his EOBs, they can be hard enough to understand that the medical identity theft might not even be noticed). Rather, the immediate victim is the insurer, who pays for care for someone who did not buy insurance. And if the insurer discovers the identity theft, the care provider becomes the victim, since the insurer may try to recover the funds paid to the provider for the imposter's care.
Unlike a stolen credit card number, which can be used to purchase almost anything (including cash cards), a stolen medical identity is not as easy to immediately monetize. However, the lower level of vigilance by the potential victim makes medical identity theft easier to pull off.
More importantly, however, the risks of medical identity theft far outweigh the risk of credit card theft or regular identity theft. An impostor who receives care while posing as the insured will leave behind a medical record that might be relied upon by some future healthcare provider. Perhaps the impostor is not allergic to penicillin, but the insured is; the impostor receives care at a hospital and the medical record says the patient may have penicillin. When the real insured shows up, tragedy might occur. Thus, while regular identity theft might cause financial ruin to its victims, medical identity theft can kill.
Does the Anthem hack indicate that an epidemic of medical identity theft is on its way? Most criminals are looking for quick cash, and medical identity theft doesn't offer as quick a reward as access to a bank account or credit card number. However, given that there is profit to be made in medical identity theft, and the risks are much greater, healthcare providers, insurers, and patients should all be on high alert for signs of it, and be prepared to quickly respond.
Anthem Breach: Secondary Impacts on Employers.
Jeff [10:06 AM]
One thing to think about when you hear of big insurers being subject to a data breach: in many cases, while the company usually does have a great deal of insured beneficiaries (either through direct insurance purchases or fully-insured employers), almost all have a great many more beneficiaries covered as TPAs or otherwise. For example, most Americans with private insurance are insured by employers who have self-funded insurance plans. Those self-finded insurance plans then go and hire Anthem, United Healthcare, Blue Cross Blue Shield, Cigna, Aetna, or some other entity to administer those plans, and those third-party administrators (or TPAs) are usually insurance companies themselves; that makes sense, since they must know how to administer the employer's self-funded plan if they can administer their own insurance products.
So, when an insurer like Anthem suffers a breach, many of the impacted individuals will be direct Anthem subscribers, but more will likely be beneficiaries of some employer who hired Anthem as a TPA of its self-insured plan.
Thus, in addition to pondering Anthem's fate, and what Anthem ought to do, it makes sense to also ponder what those self-insured plans and plan sponsors ought to do. Interestingly, here's an employment law boutique with a blog post
on just that. Something for employer clients of Anthem to consider, for sure, and useful thoughts for all employers with either fully-insured or self-insured/TPA plans. Additionally, it's worth it for employers to start thinking about what they would do if such a breach occurred with their own TPA.
Update: Here's another (shorter) blog post
with an additional good point: check your BAAs to see who is responsible for notifications. Of course, if you are (i) a HIPAA covered entity or (ii) a HIPAA business associate with any possible breach notification obligations, you should already have breach notification communication tools (set channels of communication, form letters, vendors chosen if not actually lined up, etc.) in place, ready to pick up and use.
[ Thursday, February 05, 2015 ]
Jeff [2:22 PM]
HIPAA for Paralegals Webinar: if you're a paralegal interested in how HIPAA works, why providers hesitate to give you medical records you've requested for litigation purposes, or how to get those covered entities to give you those records, you might want to check out this webinar
I'm putting on next week. You can get a 50% discount if you use priority code 15999 and discount code O7839374.
Jeff [11:12 AM]
By now you've heard of the latest huge data breach
. Just a reminder, PHI has value as a breach target on multiple fronts: direct theft of account numbers (particularly credit card numbers that can be used immediately), regular identity theft value (stealing SSNs to get credit cards or loans in the victim's name), medical identity theft (to pose as the victim and use up their insurance benefits), sensitivity/"hostage" value (to obtain information on particular individuals for extortive uses or to extort the covered entity), etc. It appears that Anthem's credit card info was protected (probably in accordance with PCI standards), but the other PHI also has value.
[ Tuesday, January 27, 2015 ]
Reporting Breaches of Less Than 500 Individuals
Jeff [12:50 PM]
: Don't forget that for "small" breaches (those involving less than 500 people), even though you need not report to OCR at the same time you report to the patient, you must still report to OCR during the first 2 months of the next calendar year. We are about halfway through that reporting period, so don't forget to log those minor breaches.
In other words, if you sent a breach notification to anyone in 2014, and did not at the same time notify OCR, then you need to do so now. You may have sighed with relief that you did not need to notify OCR (and the media) at the time, and your notification now will not lead to a "wall of shame" posting, but you must still notify OCR.
You can do so by going here
and following the link to "Breaches Affecting Fewer Than 500 Individuals."
[ Monday, January 26, 2015 ]
Jeff [11:44 AM]
Jeff [11:36 AM]
How does HIPAA apply to health apps? It seems pretty easy to me (follow the trail from the covered entity -- plan, provider, or clearinghouse -- to see if the app provider is a BA, subcontractor BA, etc.), but apparently there's some confusion
in the industry, since OCR has indicated they will give guidance soon.
[ Sunday, January 25, 2015 ]
New Jersey Requires Encryption:
Jeff [4:38 PM]
Beginning August 1, Garden State insurers
and healthcare providers must now encrypt all PHI
they collect or possess. It's more restrictive than HIPAA (where encryption is not required but is an addressable standard) so it's not preempted. This will raise issues for multi-state providers and insurers.
UPDATE: the new New Jersey bill only applies to "health insurance carriers," not to providers. The blog post I linked to implied that healthcare providers were also covered. Not so.
Hat tip: Theresa Defino
[ Tuesday, January 20, 2015 ]
Healthcare.gov: Who Has Access to What Data?
Jeff [1:56 PM]
It seems a lot of vendors can access information
of people who log onto healthcare.gov to look for Obamacare insurance or information. I wonder if these vendors are subject to BAAs or other privacy and confidentiality restrictions. I don't believe the website is a HIPAA covered entity. . . .
Is HIPAA Enforcement Increasing?
Jeff [1:54 PM]
Or is the delay in the roll-out of the 2nd phase of audits indicative of lackluster enforcement by OCR? That's what some are saying
[ Monday, January 12, 2015 ]
Bleg: I'm looking for recommendations for HIPAA auditors, to review a covered entity for HIPAA compliance. Let me know if you have any recommendations.
Jeff [3:16 PM]
Jeff [2:06 PM]
Jeff [1:06 PM]
[ Friday, January 09, 2015 ]
Bad Employee Does Not Always Equal Employer Liability
Jeff [4:30 PM]
: Kettering Health Network was sued
by a woman whose estranged husband, while an employee of Kettering, wrongfully accessed her and family members' medical records. She sued on a qui tam action, alleging the hospital wrongfully took Meaningful Use funds. The court ruled against her, saying Kettering in fact did what they were supposed to under Meaningful Use: they installed controls and did the risk analysis. The controls obviously weren't fool-proof, cause that fool husband got around them. But it was actually the hospital that caught him, proving that the controls were at least of some use.
This is not exactly opposite of the Indiana Walgreens case (there are tons of differentiating factors), but indicates that just because an employee goes rogue, it doesn't necessarily mean the employer has to be liable.
[ Monday, January 05, 2015 ]
Medical Reality TV Shows and Potential HIPAA Violations:
Jeff [11:54 AM]
Interesting (but obviously one-sided -- it is the New York Times) article on a family's lawsuit
against a hospital for allowing seemingly de-identified PHI to be used in a reality TV show. The patient's face was blurred, but the family was able to tell it was him.
[ Tuesday, December 30, 2014 ]
Jeff [1:44 PM]
[ Tuesday, December 23, 2014 ]
Jeff [1:13 PM]
2014 OCR Enforcement Actions:
Jeff [1:01 PM]
Ed Zacharias runs down OCR's year of settlements
. I tend to agree that enforcement will continue to increase, and that having good, regularly updated policies and procedures (based on a valid and thorough risk assessment), and following them, might not completely insulate you, but will go a long way to keeping fines down.
Blogger: HIPAA Blog - Edit your Template