Health knowledge made personal
Join this community!
› Share page:
Search posts:

“Distressed Babies” and Employment-Based Health Coverage

Posted Feb 11 2014 12:00am
jacobi-176x220_1 AOL CEO Tim Armstrong was the center of a story bringing together issues of whining one-percenters, employees’ rights to privacy in their health information, and the growing evidence that some employers are simply not good agents for their employees’ health coverage.

 Armstrong is an accident-prone speaker . Most recently, he gained attention for the way he explained a change in AOL’s methods of contributing to employees’ 401(k) plans. AOL announced that I planned to shift from a system by which employees received matches for their contributions each pay period, to one in which the company contributes only at the end of the year. AOL has been criticized for this move, as it deprives employees who move on during the year of any contributions, and denies ongoing employees a substantial amount of the money’s time value.  AOL has had second thoughts , and apparently will not change its current 401(k) policy.

But Armstrong’s curious explanations for the harsh move generated more interest than the change itself, and the story therefore lives on. After initially explaining that Obamacare made him do it, a published informal transcript of his remarks to employees has him shifting the blame in part to the cost of care for two mothers and children covered by AOL’s employee health plan:

[W]e had two AOLers that had distressed babies that were born that we paid a million dollars each to make sure those babies were okay in general.

It is not clear from the reports what Armstrong meant by “distressed babies.” Also unexplained was how this particular health cost so deeply affected AOL, with over 5,000 employees , $2.3 Billion in gross revenue in 2013 (and second on Fortune’s list for profit growth ), and with a CEO pulling down over $12 Million . Of course some insureds use more health care than others in any year; that’s why the cost-spreading mechanism of health insurance is so important. Or, as Deanna Fei, identifying herself in Slate as the mother of one of the babies, said,

[W]e experienced exactly the kind of unforeseeable, unpreventable medical crisis that any health plan is supposed to cover. Isn’t that the whole point of health insurance?

In short, the message is that two covered families suffered catastrophic childbirth experiences last year, and those experiences justified a reduction in retirement benefits. Really. But as even the WSJ noted , employees have privacy rights in their medical information. And the vivid description of the employees’ experiences may well constitute the release of Protected Health Information (PHI), leading to the families and their medical experiences being outed to coworkers and others.

AOL may be self-insured, in which case it operates a health plan that may be a covered entity for HIPAA purposes. And/or, AOL may have a Business Associate relationship with an insurer, third party administrator, or other covered entity. It appears, in any event, that Armstrong may have received PHI regarding these two employees – and then felt free to use the PHI as a talking point. AOL and/or its plan may have to deal with charges of violation of the following HIPAA regulation:

[With some exceptions,] a group health plan, in order to disclose protected health information to the plan sponsor or to provide for or permit the disclosure of protected health information to the plan sponsor by a health insurance issuer or HMO with respect to the group health plan, must ensure that the plan documents restrict uses and disclosures of such information by the plan sponsor consistent with the requirements of this subpart.

45 CFR 164.504(f)(1)(i). It is unlikely that disclosing the information to justify cutting back on pension benefits is a permitted disclosure.

In addition, the ADA (see 42 USC § 12112(d)) limits the use of medical information by employers – whether or not the employee is a person with a disability. The EEOC states [  ] the general rule well:

Medical records are confidential. The basic rule is that with limited exceptions, employers must keep confidential any medical information they learn about an applicant or employee.

So, Armstrong’s discussion of his employees’ medical condition might violate the law. The tone of Armstrong’s message with respect to his employees’ benefits also raises an issue I’ve touched on in a previous post . Briefly stated, the compromises leading to the enactment of the Affordable Care Act left employment-based health coverage at the center of our health care system. Most employers are good stewards of their employees’ health coverage, and make decisions in the employees’ best interest. Some do not embrace the role of faithful agent for their employees in such matters – far from it. As we work to improve the implementation of the ACA, we should be mindful that there were and are other ways to organize our health insurance system, many of which would take the Tim Armstrongs of the world out of the position of making decisions about Americans’ health care, divesting him of the power to disclose PHI on a whim, and perhaps protecting his employees from his next cut – perhaps of their health benefits.

1. At Wonkblog, Lydia DePillis comments on a new “partnership of breathtaking scope” between the National Institutes of Health and ten drug companies.  She writes:  “It took NIH Director Francis Collins more than two years to wrangle the industry into cooperating. They brought in consultants to structure the $230 million project, which will be jointly funded by companies and the government; several companies and one mental disorder schizophrenia fell by the wayside in the process. It ended up being a remarkable arrangement for a sector that’s long been fiercely competitive, jealously guarding its intellectual property against would-be copycats: All the results they come up with will be open source. At the end of the project, which should lock in a basic understanding of how the diseases [Alzheimer's, Type 2 diabetes, lupus and rheumatoid arthritis] work, they’ll race to bring new drugs to market as fast as possible.”

2. Katie Thomas of the New York Times reports on emails recently made public as a result of products liability suits that suggest that “[t]he makers of the blood-thinning drug Pradaxa were so worried that an internal research paper would damage drug sales that some employees not only pressured the author to revise it, but suggested it should be quashed altogether[.]“

3. Health Affairs publishes a fascinating-looking issue “largely devoted to papers that report early evidence and future promise of ‘connected health,’ the umbrella term arrived at to lessen the confusion over the definitions of telemedicine, tele-health, and mHealth.”

4. At JAMA, Troyen Brennan, Chief Medical Officer at CVS Caremark, and Steven Schroeder wrote a Viewpoint piece on CVS’ decision to stop selling cigarettes and other tobacco products.  They explain: “if people understand that retail outlets that plan to promote health, provide pharmacy services, and house retail clinics are no longer going to sell tobacco products, the social unacceptability of tobacco use will be substantially reinforcedindeed, the continued sale would appear to sanction the most unhealthy habit a person can maintain. If pharmacies do not make this effort voluntarily, federal or state regulatory action would be appropriate.”

5. Finally, at the FDA Law Blog, Jeffrey Wasserstein and Jennifer Thomas take note of a seeming contradiction between HHS’ recently-released final rule giving patients direct access to the results of laboratory tests, on the one hand, and the warning letter FDA sent to 23andMe late last year, on the other.  Wasserstein and Thomas conclude that “the questions raised by these two government actions remain: when does greater access to health information empower patients, rather than putting them at risk?, or, when is patient empowerment worth the risk? The answer for now seems to be ‘when HHS and FDA say so.’”

Donna Hanrahan_Headshot (2) Online social networks have potential to change the nature, speed, and scope of public health surveillance and research by offering a real-time stream of user-generated updates from millions of people around the world. In recent years, systems using informal data mined from social media sources have been credited with reducing the time it takes to detect an emerging outbreak , preventing governments from suppressing outbreak information , facilitating public health responses , and contributing to health risk behavior research  in a quick and cost-efficient manner.

Despite the inherent public nature of social media, there are many ethical implications inherent in the systematic acquisition of personal information, especially that pertaining to health. Concerns surrounding social network data analysis include issues of privacy, data quality, public panic, autonomy, access, and informed consent. While online social network data analysis holds great promise, it is essential that this valuable data be systematically harnessed in compliance with the law and ethical principles to yield population-level health benefits.

Privacy Concerns

Advancements in information and communication technologies distort the boundaries between what is public and what is private. Users of online social networks often share identifiable information about themselves, including their full names, birthdates, email addresses, GPS coordinates, and job titles. By providing researchers with rich, ready-made data sets, social media is incentivizing researchers to develop innovative methods to search the Internet for health-related information. The mining and mapping of social networks have become a common practice, from market research to medical studies. However, it is important then to consider what obligations researchers and public health officials have in meeting their online subjects’ expectations of privacy.

The Code of Federal Regulations governing human subject research, 45 C.F.R. § 46.102 , defines private information as individually identifiable information about behavior “that occurs in a context in which an individual can reasonably expect that no observation or recording is taking place, and information which has been provided for specific purposes by an individual and which the individual can reasonably expect will not be made public.”

While mining publicly available data from open sources is within the letter of the law, it raises a number of ethical issues. Some might argue it seems unreasonable that a public posting on a public site can hold an expectation of privacy. However, privacy can conceptually be considered to be an individual’s right to determine what information one would like to share with others and the ability to control when others can access that information. While the practice of data mining is growing, many social media users are unaware of how public their data truly is.

Researchers must take into consideration the level of sensitivity of the information detected, such as stigmatized health conditions. Recent studies have shown that the Internet is used more often to get health information by patients with stigmatized conditions, such as mental disorders and sexually transmitted diseases.  The misuse of such data collected from the Internet by researchers can lead to stigma, discrimination, and discomfort of the subject. With no national standard, researchers and bioethicists are left to grapple with the issue of determining what situations render it permissible to turn unsuspecting individuals into a research subjects without notification or informed consent. Historically, advancements in bioethics standards have been reactionary to human subject abuses. It is vital to resist this reactionary approach to the lack of oversight in internet research and take a proactive stance to develop acceptable standard operating procedures for the use of big data sets culled from online social network websites before foreseeable abuses occur.


Privacy concerns notwithstanding, the potential societal benefit of digital epidemiology remains clear. The use of social media has the capacity to transform disease surveillance and change how healthcare workers respond to public health emergencies. As public health threats become increasingly complex, trade-offs must be made to ensure the collective benefits of population health warrant infringement on individual rights, while balancing competing ethical, health, economic, and legal concerns. Public health researchers must work together with policy makers, medical professionals, and bioethicists to develop unambiguous ethical guidelines to answer to challenges stemming from today’s technological advances and changing communications structure.

For more on this topic, visit .

Kate Greenwood_high res 2011 comp

Cross-Posted at Bill of Health

Last week, ran an interesting article by Laura Krantz on the difficulties pregnant women and new mothers who are addicted to drugs have accessing not just drug treatment but also all of the other services and supports they need. Krantz reported on a hearing before the Human Services Committee of the Vermont House of Representatives at which a new mother in recovery from addiction, “a neonatalogist, a substance abuse clinician, a Health Department employee and a representative from the Phoenix House, a residential treatment facility in Brattleboro … all said women need not only treatment, but housing, transportation and help finding jobs.”

Alice Larned, a substance abuse clinician at the Lund Family Center in Burlington, told Krantz that spaces in residential detoxification facilities are increasingly scarce. The demand for transitional housing for women who have completed inpatient detoxification also exceeds the supply. Add to this the sad fact that women can wait a year or more for an appointment with a physician who can treat them with methadone or buprenorphine. Larned told Krantz that many of the women who start treatment with her are taking buprenorphine they bought illegally, an “indication they want help ‘yet we don’t have the legitimate means for them to get this medication[.]’”

In another story that ran last week on NPR, Steve Zind spoke with Harry Chen, the Commissioner of the Vermont Department of Public Health, who emphasized the complexities inherent in treating addiction in pregnant women and new mothers. To do so successfully, Commissioner Chen explained, “requires so many different systems working together well: the social service system, the health care system, the substance abuse system and even to some extent the correctional system.”

I confess that one reason that these two articles caught my attention is that Alice Larned is my sister.  Another reason, though, is that the problem described in the articles seems like a promising application for social impact financing, something that has been in the news here in New Jersey in recent weeks.

1. At Regulatory Reports, Alec Gaffney wrote an article, At 160 Million Patients, FDA’s Mini-Sentinel Isn’t so ‘Mini’ Anymore , in which he reports that “[a]s of July 2012, the Mini-Sentinel System’s Distributed Database (MSDD) includes information on 160 million individuals, 3.5 billion medication dispensingsmore than 45 million per monthand 3.8 billion unique medical encounters.”  According to Gaffney, Mini-Sentinel can already provide the FDA “with key information on the safety of products far more quickly than the agency could do by itself in the past. … In other words, thanks to big data, the only thing ‘mini’ about the Sentinel system may wind up being the waiting time between regulators asking a question and receiving an answer.”

2. An article in JAMA Internal Medicine, Nudging Guideline-Concordant Antibiotic Prescribing: A Randomized Clinical Trial , gives the results of a clinical trial that ”found a significant decrease in unnecessary antibiotic prescribing rates for patients treated by clinicians who signed and posted a letter in their examination rooms emphasizing a commitment to avoid inappropriate antibiotic prescribing for [acute respiratory infections].”  The authors comment that interventions grounded in a rational model of clinician behavior, such as education, electronic alerts, and financial incentives, ”have not been particularly effective.” They call for investigation of novel approaches, such as theirs, that appeal to psychological factors, such as our strong motivation to follow through on our publicly-made commitments.

3. On Thursday, January 30th, Dr. Jeffrey Shuren of the Food and Drug Administration’s Center for Devices and Radiological Health said at an event sponsored by the Pew Charitable Trusts that the FDA believes that there should be an expedited pathway to approval for high-need devices, paired with agency authority to order post-approval studies of devices, authority it already has with regard to drugs.  David Pittman reports on Shuren’s speech at MedPage Today, here .

4. Also on Thursday, CMS announced “interim financial results for select Medicare Accountable Care Organization (ACO) initiatives, an in-depth savings analysis for Pioneer ACOs, results from the Physician Group Practice demonstration, and expanded participation in the Bundled Payments for Care Improvement Initiative.  Savings from both the Medicare ACOs and Pioneer ACOs exceed $380 million.”  While less than half of the Medicare ACOs (54 of 114) that started in 2012 succeeded in cutting costs, CMS emphasized that the goal is “to achieve savings over several years, not always on an annual basis[.]“  The Pioneer ACOs also had inconsistent performance, with just 9 of 23 succeeding in cutting costs while maintaining quality.  The Pioneer ACOs nonetheless generated savings that ”far exceed findings from a previous analysis conducted by CMS, which used a different methodology.”

5. Finally, I highly recommend this  article  from last week by Jenny Gold at KHN News about how parents sharing information with other parents changed the standard of care for treating clubfoot.  Here’s a taste: “Surgeons are trained to operate…and usually that’s the way they make money. The Ponseti Method brings in a lot less for orthopedists. For about 50 years, technique mostly stayed in Iowa.  But then something new came along: The internet.”

Post a comment
Write a comment: