This season, “Downton Abbey's” plot line has health policy wonks on the edge of their seats: a heated debate about hospital consolidation that closely parallels what’s going on in the U.S. health care system today.

If you’re not a Downton fan, here’s a quick plot recap by Kaiser Health News reporter Jenny Gold: It’s 1925 for the lords and ladies at Downton Abbey. Think flapper dresses, cocktail parties and women’s rights. And a big hospital in the nearby city of York is making a play to take over the Downton Cottage Hospital next to the posh estate.

As Maggie Smith’s character, the Dowager Countess of Grantham, sees it, “The Royal Yorkshire county hospital wants to take over our little hospital, which is outrageous!”

Stanford Health Policy’s Kathy McDonald — an unabashed fan of the popular PBS period piece — says things haven’t changed that much today. There has been an uptick in hospital consolidations since 2010, with about 100 taking place each year, she says.

You can listen to McDonald’s interview with Gold, who took the Downton debate to the American Public Media radio show, “Marketplace.”

Image

A substantial share of all malpractice claims in the United States is attributable to a small number of physicians, according to a study led by researchers at Stanford University and the University of Melbourne.

The team found that just 1 percent of practicing physicians accounted for 32 percent of paid malpractice claims over a decade. The study also found that claim-prone physicians had a number of distinctive characteristics, such as practicing specialities that are riskier than others.

“The fact that these frequent flyers looked quite different from their colleagues — in terms of specialty, gender, age and several other characteristics — was the most exciting finding,” said David Studdert, professor of medicine and of law at Stanford. “It suggests that it may be possible to identify high-risk physicians before they accumulate troubling track records, and then do something to stop that happening.”

Studdert is also a core faculty member at Stanford Health Policy and the lead author of the study published in The New England Journal of Medicine.

Concentrated among a small group

“The degree to which the claims were concentrated among a small group of physicians was really striking,” added Studdert, an expert in the fields of health law and empirical legal research.

The researchers analyzed information from the U.S. National Practitioner Data Bank, a data repository established by Congress in 1986 to improve health-care quality. Their study covered 66,426 malpractice claims paid against 54,099 physicians between January 2005 and December 2014.

Almost one-third of the claims related to patient deaths; another 54 percent related to serious physical injury. Only 3 percent of the claims were litigated to verdicts for the plaintiff. The remainder resulted in out-of-court settlements. Settlements and court-ordered payments averaged $371,054.

“The concentration of malpractice claims among physicians we observed is larger than has been found in the few previous studies that have looked at this distributional question,” said Michelle Mello, a co-author of the study and professor of law and of health research and policy at Stanford.

“It’s difficult to say why that is,” Mello added. “The earlier estimates come from studies of single insurers or single states, whereas ours is national in scope. Also, the earlier numbers are more than 25 years old now, and claim-prone physicians may be a bigger problem today than they were then.”

Encouraging greater awareness

The authors recommend that all institutions that handle large numbers of patient complaints and claims develop a greater awareness of how these events are distributed among clinicians.

“In our experience, few do,” they write in the paper. “With notable exceptions, fewer still systematically identify and intervene with practitioners who are at high risk for future claims.”

The most important predictor of incurring repeated claims was a physician’s claim history. Compared to physicians with only one prior paid claim, physicians who had two paid claims had almost twice the risk of another one; physicians with three paid claims had three times the risk of recurrence; and physicians with six or more paid claims had more than 12 times the risk of recurrence.

“Risk also varied widely according to specialty,” the authors noted. “As compared with the risk of recurrence among internal medicine physicians, the risk of recurrence was approximately double among neurosurgeons, orthopedic surgeons, general surgeons, plastic surgeons and obstetrician-gynecologists.”

The lowest risks of recurrence occurred among psychiatrists and pediatricians.

Male physicians had a 40 percent higher risk of recurrence than female physicians, and the risk of recurrence among physicians younger than 35 was about one-third the risk among their older colleagues, the study found.

“If it turns out to be feasible to predict accurately which physicians are going to become frequent flyers, that is something liability insurers and hospitals would be very interested in doing,” Studdert said.

“But institutions will then face a choice,” he added. “One option is to kick out the high-risk clinicians, essentially making them someone else’s problem. Our hope is that the knowledge would be used in a more constructive way, to target measures like peer counseling, retraining, and enhanced supervision. These are interventions that have real potential both to protect patients and reduce litigation risks.”

Other stories about the study:

The New York Times Well Blog

The Huffington Post

KQED Public Radio

U.S. News & World Report

CBS News

Reuters

Medscape

The first Liberation Technology seminar of the winter quarter on January 6, featured Jonathan Zittrain, Professor of Law at Harvard Law School and co-founder of the Berkman Center for Internet & Society at Harvard University. Zittrain focused his talk entitled, Minds for Sale, on the variety of online platforms that harness the wisdom of crowds today, and closed with a discussion of the implications of these platforms. Categorizing these new tools by the breadth of their user base, Zittrain began by describing the platforms that pay the most money per task, require the most skill, and subsequently have the least participation. Key examples of these platforms are listed below, and organized by their decreasing level on skill.

•  Xprize Foundation attempts to prompt radical breakthroughs through competitions for large quantities of prize money.
• InnoCentive creates a marketplace between engineers and scientists to encourage innovation.
• LiveOps, which bills itself as a contact center cloud, has developed a large set of independent contractors who are designated specific tasks when they sign in to the site. They may be assigned to answer calls placed to a restaurant or emergency hotline or to make political calls on behalf of Liveops clients.
• Samasource offers "dignified digital work for women, children and refugees" located in developing countries.
• Amazon Mechanical Turk (also known as Artificial Artificial Intelligence) allows users to participate in anonymous, minimally paid tasks. These tasks can be submitted by any party and are highly disaggregated among hundreds or thousands of users, each of whom receives a payout of between approximately one to fifteen cents for completing the task.
• Soylent, which calls itself "a word processor with a crowd inside" embeds workers from Mechanical Turk into Microsoft Word. When users install the site's Shortn add-in into Microsoft Word, they can use the tool to have written samples shortened in two minutes. In order to achieve this, the written sample is disaggregated by paragraphs, and each paragraph is shortened within two minutes by a "Turker"-a user who accepts the task on Mechanical Turk.
• Microtask enables disaggregating previously sensitive data to be work processed from a scanned image. The ultimate goal of this group is to put distributed work into video and computer games. For now, they disaggregate larger tasks into two-second tasks that can be distributed across a crowd.
• Games With A Purpose (GWAP) has a game-like platform designed to get users to complete disaggregated tasks for virtual points, rather than for actual remuneration. One of its more popular activities quickly attracted 23,000 players who contributed 4.1 million labels to images they were presented with in The ESP Game. Many players routinely play more than 20 hours a week.

Next, Zittrain moved on to offer some additional examples, hypothetical scenarios and questions that highlight some of his own concerns about the potential of these technologies. One key question is whether this market may be too efficient at linking up solvers and payers, if certain tasks are not in fact beneficial for society. After all, it was highly contentious when Texas governor Rick Perry set up video cameras on Texas' border with Mexico and invited people to watch the video feeds and report if they saw anything suspicious. Similarly, a site called InternetEyes allows users to watch video feed from CCTV cameras in the UK for free, offering the chance to "earn reward money, have a chance at reducing crime, and potentially become a hero and save lives," instead of actual compensation.

In another example, the University of Colorado Police Department recently offered a $50 reward for identification of students shown to be smoking marijuana in photos from a large April 20 gathering. Hypothetically, the Iranian government could use Turkers to identify Iranian protestors through national ID photographs in the very same way; by Zittrain's calculations, this task could be disaggregated and carried out quickly at a cost of only $17,000 per protester identified.

Another cluster of issues relates to the use of anonymous users to carry out disaggregated tasks that may have the effect of exerting influence on others. For example, Turkers were recently offered the opportunity to write a positive 5 out of 5 review for a product on a website, which hundreds accepted and completed for a few cents at a time. Users on SubvertandProfit.com are paid to "Like" something on Facebook, "Digg" something, or show their approval of a site or product via some other social network. The users are remunerated for their effort, and the site also profits. Taking this a step further, these platforms also enable users to pay people to evince opinions on legislative issues they do not actually care about. For example, health insurers were recently caught paying Facebook gamers virtual currency to oppose the health reform bill. By enabling any task to be disaggregated and monetized, these new platforms can have highly controversial and unethical implications.