Category Archives: data saving lives

complex adaptive systems, data saving lives, healthcare, rapid learning health systems

NYT: Can Big Data Tell Us What Clinical Trials Don’t?

 

http://www.nytimes.com/2014/10/05/magazine/can-big-data-tell-us-what-clinical-trials-dont.html?src=twr

MAGAZINE

Can Big Data Tell Us What Clinical Trials Don’t?

Illustration by Christopher Brand

Eureka

By VERONIQUE GREENWOOD

When a helicopter rushed a 13-year-old girl showing symptoms suggestive of kidney failure to Stanford’s Packard Children’s Hospital, Jennifer Frankovich was the rheumatologist on call. She and a team of other doctors quickly diagnosed lupus, an autoimmune disease. But as they hurried to treat the girl, Frankovich thought that something about the patient’s particular combination of lupus symptoms — kidney problems, inflamed pancreas and blood vessels — rang a bell. In the past, she’d seen lupus patients with these symptoms develop life-threatening blood clots. Her colleagues in other specialties didn’t think there was cause to give the girl anti-clotting drugs, so Frankovich deferred to them. But she retained her suspicions. “I could not forget these cases,” she says.

Back in her office, she found that the scientific literature had no studies on patients like this to guide her. So she did something unusual: She searched a database of all the lupus patients the hospital had seen over the previous five years, singling out those whose symptoms matched her patient’s, and ran an analysis to see whether they had developed blood clots. “I did some very simple statistics and brought the data to everybody that I had met with that morning,” she says. The change in attitude was striking. “It was very clear, based on the database, that she could be at an increased risk for a clot.”

The girl was given the drug, and she did not develop a clot. “At the end of the day, we don’t know whether it was the right decision,” says Chris Longhurst, a pediatrician and the chief medical information officer at Stanford Children’s Health, who is a colleague of Frankovich’s. But they felt that it was the best they could do with the limited information they had.

A large, costly and time-consuming clinical trial with proper controls might someday prove Frankovich’s hypothesis correct. But large, costly and time-consuming clinical trials are rarely carried out for uncommon complications of this sort. In the absence of such focused research, doctors and scientists are increasingly dipping into enormous troves of data that already exist — namely the aggregated medical records of thousands or even millions of patients to uncover patterns that might help steer care.

The Tatonetti Laboratory at Columbia University is a nexus in this search for signal in the noise. There, Nicholas Tatonetti, an assistant professor of biomedical informatics — an interdisciplinary field that combines computer science and medicine — develops algorithms to trawl medical databases and turn up correlations. For his doctoral thesis, he mined the F.D.A.’s records of adverse drug reactions to identify pairs of medications that seemed to cause problems when taken together. He found an interaction between two very commonly prescribed drugs: The antidepressant paroxetine (marketed as Paxil) and the cholesterol-lowering medication pravastatin were connected to higher blood-sugar levels. Taken individually, the drugs didn’t affect glucose levels. But taken together, the side-effect was impossible to ignore. “Nobody had ever thought to look for it,” Tatonetti says, “and so nobody had ever found it.”

The potential for this practice extends far beyond drug interactions. In the past, researchers noticed that being born in certain months or seasons appears to be linked to a higher risk of some diseases. In the Northern Hemisphere, people with multiple sclerosis tend to be born in the spring, while in the Southern Hemisphere they tend to be born in November; people with schizophrenia tend to have been born during the winter. There are numerous correlations like this, and the reasons for them are still foggy — a problem Tatonetti and a graduate assistant, Mary Boland, hope to solve by parsing the data on a vast array of outside factors. Tatonetti describes it as a quest to figure out “how these diseases could be dependent on birth month in a way that’s not just astrology.” Other researchers think data-mining might also be particularly beneficial for cancer patients, because so few types of cancer are represented in clinical trials.

As with so much network-enabled data-tinkering, this research is freighted with serious privacy concerns. If these analyses are considered part of treatment, hospitals may allow them on the grounds of doing what is best for a patient. But if they are considered medical research, then everyone whose records are being used must give permission. In practice, the distinction can be fuzzy and often depends on the culture of the institution. After Frankovich wrote about her experience in The New England Journal of Medicine in 2011, her hospital warned her not to conduct such analyses again until a proper framework for using patient information was in place.

In the lab, ensuring that the data-mining conclusions hold water can also be tricky. By definition, a medical-records database contains information only on sick people who sought help, so it is inherently incomplete. Also, they lack the controls of a clinical study and are full of other confounding factors that might trip up unwary researchers. Daniel Rubin, a professor of bioinformatics at Stanford, also warns that there have been no studies of data-driven medicine to determine whether it leads to positive outcomes more often than not. Because historical evidence is of “inferior quality,” he says, it has the potential to lead care astray.

Yet despite the pitfalls, developing a “learning health system” — one that can incorporate lessons from its own activities in real time — remains tantalizing to researchers. Stefan Thurner, a professor of complexity studies at the Medical University of Vienna, and his researcher, Peter Klimek, are working with a database of millions of people’s health-insurance claims, building networks of relationships among diseases. As they fill in the network with known connections and new ones mined from the data, Thurner and Klimek hope to be able to predict the health of individuals or of a population over time. On the clinical side, Longhurst has been advocating for a button in electronic medical-record software that would allow doctors to run automated searches for patients like theirs when no other sources of information are available.

With time, and with some crucial refinements, this kind of medicine may eventually become mainstream. Frankovich recalls a conversation with an older colleague. “She told me, ‘Research this decade benefits the next decade,’ ” Frankovich says. “That was how it was. But I feel like it doesn’t have to be that way anymore.”

data saving lives, health market quality, healthcare, policy, politics

Health Data “Interoperability”: A $30 Billion Unicorn Hunt

too funny

http://www.forbes.com/sites/theapothecary/2014/09/03/health-data-interoperability-a-30-billion-unicorn-hunt/

Health Data “Interoperability”: A $30 Billion Unicorn Hunt

John R. GrahamJohn R. Graham , Contributor
Comment Now

Having cheered as $26 billion of taxpayers’ money has been spent since 2009 inducing hospitals and physicians to install electronic health records (EHRs), many champions of the effort are dismayed that the EHRs are not interoperable. That is, they cannot talk to each other – which was the whole point of subsidizing the exercise.

All this money has achieved a process goal: There has been a significant uptake of EHRs. According to a recent review, the proportion of physicians who have at least a basic EHR has increased from under 22 percent to 48 percent. Doctors were motivated by the bounty offered, plus the threat of having reimbursements being clawed back in 2015 if they have not adopted EHRs. The proportion of hospitals has similarly increased from 12 percent to 44 percent.

But these EHRs do not  talk to each other. According to the same review, “only 10 percent of ambulatory practices and 30 percent of hospitals were found to be participating in operational health information exchange efforts.”

All those billions of taxpayers’ dollars were paid out to providers who attest to “meaningful use” of EHRs. However, there are three stages of meaningful use.  Stage 1 was relatively simple. Stage 2 was originally supposed to be achieved by 2013, but that has been pushed back until 2016. The hang up is that stage 2 has a high hurdle for interoperability.

According to the final rule published in September 2012, requirements include “the expectation that providers will electronically transmit patient care summaries with each other and with the patient to support transitions in care. Increasingly robust expectations for health information exchange in Stage 2 and Stage 3 would support the goal that information follows the patient.”

Despite the delay, providers are still complaining that the requirements are too demanding. According to Russell Branzell, president and CEO of the College of Healthcare Information Management Executives: “Now the very future of Meaningful Use is in question.”

So it should be: Evidence from Congressional investigations suggests that meaningful-use bounties have encouraged the adoption of EHRs that are deliberately closed to exchange with other parties. The problem is that exchanging data with competitors is fundamentally against the self-interest of the party which created the data. Nobody would expect The U.S. Department of Transportation to set up a fund to incentivize car-markers to exchange data with each other, or the U.S. Department of Agriculture to set up a fund to incentivize grocery stores to exchange data with each other.

670px-obama_signing_health_care-201003231

That is not to say that there would be no value to such data exchange. IfSafeway were out of my favorite brand of breakfast cereal, I’d love for the clerk to tell me that Giant had plenty in stock just down the road, instead of selling me something similar. However, the amount of government funding required to overwhelm competitors’ resistance to doing this would surely not be worth it.

I’m sure readers can come up with many examples that would demonstrate the public benefit of competing hospital systems sharing data seamlessly. An epidemic or terrorist attack are easy ones. However, advocates of health information exchange emphasize how it would reduce friction in the day to day operations of our health system.

But at what cost? $26 billion has not done the trick. It is unlikely that the remaining $4 billion in the pot will get the job done. The Office of the National Coordinator of Health IT has been promoting a ten-year plan for more funding – even a trust fund like the Federal Highway Trust Fund!

When the Office of the National Coordinator of Health IT was established during the Bush Administration, its purpose was to “coordinate,” not underwrite nor regulate. Congress should be wary of appropriating yet more funding to hunt the unicorn of health data interoperability.

data saving lives, facts & data points, healthy habits, nutrition, policy

The Key to Changing Individual Health Behaviors: Change the Environments That Give Rise to Them

 

http://harvardpublichealthreview.org/the-key-to-changing-individual-health-behaviors-change-the-environments-that-give-rise-to-them/

The Key to Changing Individual Health Behaviors: Change the Environments That Give Rise to Them

PDF: HPHRv2-Stulberg

Over the past four decades, the United States has faced steadily rising rates of obesity and associated chronic conditions. Many of these chronic conditions are rooted in nutrition and physical activity behaviors, and are often referred to as lifestyle diseases. Historically, the prevention of lifestyle diseases has focused on changes in individual behavior and personal choices, and personal responsibilities. However, a growing body of research has demonstrated the strong influence of physical and social surroundings on individuals’ actions. The context in which options are presented can shape the decision-making processes that impact health. Altogether, the research suggests that altering environments may be an effective driver of behavior change. 1Intentionally designing environments to promote healthy behaviors holds promise to reverse the increase of lifestyle diseases.

The emerging field of behavioral science – which gathers insights from disciplines like behavioral economics, cognitive psychology, and social psychology – illustrates that while individuals retain “free choice,” their environment significantly influences the choices they make, and in some instances, may lead them to act in ways that are counter to their true preferences. 2 A few examples:

  • Individual preferences are often inconsistent over time, especially in situations where immediate pleasures carry long term consequences. In a study that asked [hypothetically] if people would prefer fruit or chocolate as a future snack, 74% chose fruit. But, when those same participants were presented with both fruit and chocolate in real-time, 70% selected chocolate. 3
  • A person’s actions can be dramatically influenced by related contextual features. For instance, research shows that kitchenware size significantly influences serving and eating behavior. In a series of studies, individuals who were given larger bowls served themselves between 28-32% more cereal than those given smaller bowls. Studies also report that people tend to eat 90-97% of what is on their plate, irrespective of plate size. 4
  • People tend to consent to the “default option.” This has been observed in numerous situations ranging from deciding whether or not to become an organ donor to making saving allocations for retirement. For example, organ donation rates are 4% in Denmark and 12% in Germany where the default option is “opt-in.” In contrast, the rates are 86% in Sweden and nearly 100% in Austria where the default option is “opt-out.” Cultural differences cannot explain the discrepancy. 5

When these behavioral science insights are applied in the context of health, the growth of lifestyle diseases is not surprising. This expanding body of research sheds light on the difficulties of healthy living when society is dominated by the marketing of unhealthy foods and unduly large portion sizes, and where sedentary behavior is often the default option.

The good news is that the same forces that currently promote unhealthy behaviors can be used to encourage healthy ones. In their bestselling book Nudge, Richard Thaler and Cass Sunstein described “choice architecture,” or the proactive designing of environments that “nudge” people to make healthier selections while still retaining freedom of choice. 6 There are many opportunities to apply this concept to promoting healthy behaviors. In particular, given their resources, broad reach, and financial and social incentives, both governments and employers are in a unique position to promote healthy behaviors in a way that would affect many lives.

Government food programs such as the Supplemental Nutrition Assistance Program (“SNAP”) and the school lunch program could be designed to make healthy selections more accessible, and in some cases, the default options. Those that oppose the trend toward encouraging healthier foods often cite added costs and waste, arguing that children don’t like healthy foods and will throw them away uneaten. But the data tell a different story. A recent study in Childhood Obesity found that a vast majority of middle-school and high-school students like the updated and significantly healthier school lunch that was introduced in 2012. 7

Nonetheless, making the change is not cost-free. A recent meta-analysis found that the healthiest diets cost $1.50 more per-person, per-day, which amounts to $550 per-person, per-year. 8 While this amount is not insignificant, it pales in comparison to the cost of treating most diet-related chronic conditions. Designing government food programs around the “healthiest diets” may yield a positive return on investment.

Even so, many individuals – including those who do qualify for SNAP, as well as those who do not qualify for SNAP (i.e. incomes just about the SNAP cut-off) – may still struggle with affordability and availability of healthy foods. Perhaps the most sustainable and far-reaching approach to making healthy foods more accessible is to change food policies (e.g., subsidies) that currently favor the production and systematic delivery of unhealthy foods to favor healthy ones. This would likely lead to higher volumes, more efficient delivery, and lower costs for nutritious foods.

The government can also promote healthier eating by improving nutrition labeling. While the FDA’s recent proposal to ensure that serving sizes listed on food products reflect actual average consumption (e.g., nutrition specifications would reflect an entire muffin, not one-third of a muffin) is a small step in the right direction, there is potential to go a lot further. Research suggests that catchier and simplified nutrition labels could have a much greater impact on consumer behavior. 9 For example, NuVal, an independently designed system that gives food items a single overall score based on more than 30 nutrient and nutrition factors, could be considered for more widespread adoption. 10 Not only does NuVal make for easier interpretation of a product’s nutrition profile, it also enables comparison shopping between options and encourages people to “trade-up” to healthier options. 11 An additional model to consider is a traffic-light rating system that marks foods with either a green, yellow, or red light. In instances where it has already been implemented (in some private organizations and outside the United States), the traffic-light model has increased consumer awareness of health and leads to healthier purchases. 12

In addition to promoting a healthy diet, government should play an active role in encouraging physical activity through the education system (e.g., ensure the existence of meaningful recess and gym programs), transportation system (e.g., create options for safe pedestrian/bike commuting), and by supporting relevant community resources (e.g., building, maintaining, and ensuring the safety of outdoor parks and recreation centers). When options for physical activity are easily accessible, people tend to be more active. For example, a recent study published in the American Journal of Public Health illustrated that the establishment of traffic-free cycling and walking routes increased overall physical activity among those that lived nearby. 13

Employers may have the ability and incentives to move faster than government in designing health promoting environments. A healthier workforce results in both reduced health care costs and absenteeism, and in increased productivity. Recent data from the Society of Human Resource Management’s annual Employee Benefits Survey shows that employers are taking notice and increasing their investment in workforce wellness programs. While these programs have traditionally focused on offering employees classes, counseling, and incentives for healthy behaviors such as discounts on insurance premiums, subtler tweaks to the workplace itself could prove just as, if not more effective.

An example of these subtler changes is happening at Google. There, company leaders have invested in promoting employee nutrition and health. Instead of relying solely on traditional programs such as nutrition counseling and weight-loss classes, Google redesigned cafeterias to encourage healthier eating. Now, the most nutritious options are positioned at the front of the cafeteria and unhealthy foods are hidden in corners and placed in opaque bowls. Smaller plates are the norm and marked with reminder messages that “bigger dishes prompt people to eat more.” Foods are tagged with either red “warning” stickers, or green stickers signifying healthy foods. Beverage coolers stock water at eye level, and relegate sweetened beverages to the bottom where they are not as easily seen or accessed. These changes – which notably do not restrict options, but simply rearrange the way options are presented – have led to dramatic reductions in candy and sugar-sweetened beverage consumption, and increases in the use of smaller plates. 14 15

To encourage physical activity, employers can adopt similar approaches to workplace design, such as centrally located staircases and ergonomically fit workstations. Further, similar to current LEED certifications for environmentally-friendly buildings, there could also be a meaningful certification for health-promoting buildings. In addition to the design of physical workplaces, the way that work itself is conducted can also be designed to promote health. For example, some employers have made “walking meetings” a cultural norm to build physical activity into otherwise sedentary jobs. 16

 

Other Considerations

While the value of these environmental interventions is promising, there is a need for additional research that focuses on cost effectiveness. This is especially true if we hope to see increased governmental action, where broad policy implementation almost always follows a positive cost/benefit analysis. That said, some of the ideas – such as using smaller plates in government cafeterias or simplifying nutrition labels – come at relatively little additional financial cost, and have already demonstrated health-promoting benefits. These ideas could be fast-tracked for more widespread adoption.

Another potential barrier that must be overcome is the political power of special interests groups that rely on built-environments conducive to unhealthy behaviors. For example, a large part of the reason that the migration to healthier school lunches has taken so long is because various food interests have launched strong lobbying campaigns against such changes. 17 In order to transition entrenched unhealthy built-environments to healthier ones, policymakers will need to prioritize the demands of public health against the backdrop of influential and longstanding special interests

A broader approach to designing environments that promote healthy behaviors must also account for additional barriers that individuals with lower socioeconomic status commonly face. The government cannot rely solely on the private sector to drive these changes since those who stand to benefit most may be unemployed or not working for progressive employers with the resources to launch effective health campaigns. Thus, focusing on government food programs and community-based approaches that effect a lower-income demographic is critical (e.g., sidewalk coverage and safe streets, eliminating food deserts, maintaining outdoor parks). In addition to these more specific interventions, the clear connection between environment and health should only bolster the case for expanding social service programs more broadly. Realizing and addressing the fact that so many of the outcomes that lie inside of health care are rooted in factors that lie outside of health care is thus critical to improving health.

 

If we want to avert a public health crisis at the hands of chronic lifestyle-driven diseases, we need not only focus on changing individual behaviors, but also on changing the environments that give rise to those behaviors. Governments and employers must recognize the overwhelming influence of context on action, and take advantage of their unique position to intentionally shape environments that promote healthy behaviors.
  1. Kahneman, D. Thinking fast and slow. New York: Farrar, Straus, and Giroux. (2011). ↩
  2. For more on Behavioral Science, see the Behavioral Science and Policy Association and its forthcoming journal Behavioral Science and Policy. ↩
  3. Read, D., & Van Leeuwen, B. Predicting hunger: the effects of appetite and delay on choice. Organizational Behavior and Human Decision Processes. 1998; 76 (2), 189-205. ↩
  4. Van Ittersum, K., & Wansink, B. Plate size and color suggestibility: the delboeuf illusion’s bias on serving and eating behavior. Journal of Consumer Research. 2012; 39 (2), 215-228. ↩
  5. Johnson, E. J., & Goldstein, D. Do defaults save lives? Science. 2003; 302, 1338-1339. ↩
  6. Thaler, R. H., & Sunstein, C. R. Nudge: Improving decisions about health, wealth, and happiness. New York: Penguin Books. (2009). ↩
  7. Turner, L., & Chaloukpa, F. J. Perceived reactions of elementary school students to changes in school lunches after implementation of the United States Department of Agriculture’s new meals standards: minimal backlash, but rural and socioeconomic disparities exist. Childhood Obesity. 2014; 10 (4), 349-356. ↩
  8. Rao, M., Afshin, A., Singh, G., & Mozaffarian D. Do healthier foods and diet patterns cost more than less healthy options? A systematic review and meta-analysis. BMJ Open. 2013; 3 (12). doi:10.1136/bmjopen-2013-004277. ↩
  9. Roberto, C. A., & Khandpur, N. Improving the design of nutrition labels to promote healthier food choices and reasonable portion sizes. International Journal of Obesity. 2014; 38, 525-533. ↩
  10. Nuval.com. Accessed August 12, 2014. ↩
  11. Nuval.com: Trading Up Tips. Accessed August 28, 2014. ↩
  12. Sonnenberg, L., Gelsomin, E., Levy, E. D., Riis, D., Barraclough, S., & Thorndike, A., N. A traffic light food labeling intervention increases consumer awareness of health and healthy choices at the point-of-purchase. Preventative Medicine. 2013; 57 (4), 253-257. ↩
  13. Freeland, A. L., Banerjee, S. N., Dannenberg, A., L & Wendel, A. M. Walking associated with public transit: moving toward increased physical activity in the United States. American Journal of Public Health. 2013; 103 (3), 536-542. ↩
  14. Kuang, C. 6 ways Google hacks its cafeterias so Googlers eat healthier. Fast Company. April 2012; (164). ↩
  15. Wacther, Luke. Personal Interview on July 20, 2014. ↩
  16. Walking meetings could make work healthier, happier. CBS News. 07, May 2014. ↩
  17. Nixon, R. Nutrition Group Lobbies Against Healthier School Meals it Sought, Citing Cost. New York Times. 01, July 2014. ↩
data saving lives, healthy habits

Anosmia predicts longevity…

 

http://www.medicalobserver.com.au/news/noses-know-about-longevity

Noses know about longevity

A A A
3rd Oct 2014

Rada Rouse   all articles by this author

AN ELDERLY person who cannot accurately distinguish the smell of peppermint or fish may be staring death in the face, research suggests.

A study among a nationally representative sample of adults aged 57–85 found those who lost their sense of smell and were already at high risk from medical conditions had more than double the risk of dying in the next five years.

The cohort of 3000 provided baseline data by trying to identify odours, which were, in order of increasing difficulty to pinpoint, peppermint, fish, orange, rose and leather.

Five years later, the researchers assessed which participants were still alive.

Some 430, or 12.5% of the original cohort, had died.

People noted as anosmic in the first survey had a threefold increased risk of death when other factors including age, race and health were taken into consideration, the researchers said.

They noted a “dose-dependent” relationship between sense of smell and risk of death, with anosmic individuals having a greatly increased risk compared to hyposmic individuals, and the latter being more likely to die than those with a normal or “healthy” sense of smell.

The study showed 39% of anosmic individuals identified in the first test had died before the second survey. 

This compared to 19% of hyposmic people and 10% of those with a normal sense of smell.

“We believe olfaction is the canary in the coalmine of human health, not that its decline directly causes death,” the researchers wrote.

Assessment of olfactory function may be useful to help identify patients at high risk of mortality, they said.

PLoS ONE 2014; online 1 Oct

complex adaptive systems, cool, data saving lives, entrepreneurship, healthcare, management and leadership, rapid learning health systems

Artificial intelligence meets the C-suite

Artificial intelligence meets the C-suite

http://www.mckinsey.com/Insights/Strategy/Artificial_intelligence_meets_the_C-suite

Jeremy Howard: Today, machine-learning algorithms are actually as good as or better than humans at many things that we think of as being uniquely human capabilities. People whose job is to take boxes of legal documents and figure out which ones are discoverable— that job is rapidly disappearing because computers are much faster and better than people at it.

In 2012, a team of four expert pathologists looked through thousands of breast-cancer screening images, and identified the areas of what’s called mitosis, the areas which were the most active parts of a tumor. It takes four pathologists to do that because any two only agree with each other 50 percent of the time. It’s that hard to look at these images; there’s so much complexity. So they then took this kind of consensus of experts and fed those breast-cancer images with those tags to a machine-learning algorithm. The algorithm came back with something that agreed with the pathologists 60 percent of the time, so it is more accurate at identifying the very thing that these pathologists were trained for years to do. And this machine-learning algorithm was built by people with no background in life sciences at all. These are total domain newbies

 

Artificial intelligence meets the C-suite

Technology is getting smarter, faster. Are you? Experts including the authors of The Second Machine Age, Erik Brynjolfsson and Andrew McAfee, examine the impact that “thinking” machines may have on top-management roles.

September 2014

artThe exact moment when computers got better than people at human tasks arrived in 2011, according to data scientist Jeremy Howard, at an otherwise inconsequential machine-learning competition in Germany. Contest participants were asked to design an algorithm that could recognize street signs, many of which were a bit blurry or dark. Humans correctly identified them 98.5 percent of the time. At 99.4 percent, the winning algorithm did even better.Or maybe the moment came earlier that year, when IBM’s Watson computer defeated the two leading human Jeopardy! players on the planet. Whenever or wherever it was, it’s increasingly clear that the comparative advantage of humans over software has been steadily eroding. Machines and their learning-based algorithms have leapt forward in pattern-matching ability and in the nuances of interpreting and communicating complex information. The long-standing debate about computers as complements or substitutes for human labor has been renewed.

The matter is more than academic. Many of the jobs that had once seemed the sole province of humans—including those of pathologists, petroleum geologists, and law clerks—are now being performed by computers.

Sidebar

Interviewee biographies

And so it must be asked: can software substitute for the responsibilities of senior managers in their roles at the top of today’s biggest corporations? In some activities, particularly when it comes to finding answers to problems, software already surpasses even the best managers. Knowing whether to assert your own expertise or to step out of the way is fast becoming a critical executive skill.

Video

Managing in the era of brilliant machines: An interview  

Managing in the era of brilliant machines: An interview

In this interview with McKinsey’s Rik Kirkland, Erik Brynjolfsson and Andrew McAfee explain the organizational challenge posed by the Second Machine Age.

Play video

Yet senior managers are far from obsolete. As machine learning progresses at a rapid pace, top executives will be called on to create the innovative new organizational forms needed to crowdsource the far-flung human talent that’s coming online around the globe. Those executives will have to emphasize their creative abilities, their leadership skills, and their strategic thinking.

To sort out the exponential advance of deep-learning algorithms and what it means for managerial science, McKinsey’s Rik Kirkland conducted a series of interviews in January at the World Economic Forum’s annual meeting in Davos. Among those interviewed were two leading business academics—Erik Brynjolfsson and Andrew McAfee, coauthors of The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (W. W. Norton, January 2014)—and two leading entrepreneurs: Anthony Goldbloom, the founder and CEO of Kaggle (the San Francisco start-up that’s crowdsourcing predictive-analysis contests to help companies and researchers gain insights from big data); and data scientist Jeremy Howard. This edited transcript captures and combines highlights from those conversations.

The Second Machine Age

What is it and why does it matter?

Andrew McAfee: The Industrial Revolution was when humans overcame the limitations of our muscle power. We’re now in the early stages of doing the same thing to our mental capacity—infinitely multiplying it by virtue of digital technologies. There are two discontinuous changes that will stick in historians’ minds. The first is the development of artificial intelligence, and the kinds of things we’ve seen so far are the warm-up act for what’s to come. The second big deal is the global interconnection of the world’s population, billions of people who are not only becoming consumers but also joining the global pool of innovative talent.

Erik Brynjolfsson: The First Machine Age was about power systems and the ability to move large amounts of mass. The Second Machine Age is much more about automating and augmenting mental power and cognitive work. Humans were largely complements for the machines of the First Machine Age. In the Second Machine Age, it’s not so clear whether humans will be complements or machines will largely substitute for humans; we see examples of both. That potentially has some very different effects on employment, on incomes, on wages, and on the types of companies that are going to be successful.

Video

Putting artificial intelligence to work: An interview with Anthony Goldbloom and Jeremy Howard

Machine-learning experts Anthony Goldbloom and Jeremy Howard tell McKinsey’s Rik Kirkland how smart machines will impact employment.

Jeremy Howard: Today, machine-learning algorithms are actually as good as or better than humans at many things that we think of as being uniquely human capabilities. People whose job is to take boxes of legal documents and figure out which ones are discoverable— that job is rapidly disappearing because computers are much faster and better than people at it.

In 2012, a team of four expert pathologists looked through thousands of breast-cancer screening images, and identified the areas of what’s called mitosis, the areas which were the most active parts of a tumor. It takes four pathologists to do that because any two only agree with each other 50 percent of the time. It’s that hard to look at these images; there’s so much complexity. So they then took this kind of consensus of experts and fed those breast-cancer images with those tags to a machine-learning algorithm. The algorithm came back with something that agreed with the pathologists 60 percent of the time, so it is more accurate at identifying the very thing that these pathologists were trained for years to do. And this machine-learning algorithm was built by people with no background in life sciences at all. These are total domain newbies.

Andrew McAfee: We thought we knew, after a few decades of experience with computers and information technology, the comparative advantages of human and digital labor. But just in the past few years, we have seen astonishing progress. A digital brain can now drive a car down a street and not hit anything or hurt anyone—that’s a high-stakes exercise in pattern matching involving lots of different kinds of data and a constantly changing environment.

Why now?

Computers have been around for more than 50 years. Why is machine learning suddenly so important?

Erik Brynjolfsson: It’s been said that the greatest failing of the human mind is the inability to understand the exponential function. Daniela Rus—the chair of the Computer Science and Artificial Intelligence Lab at MIT—thinks that, if anything, our projections about how rapidly machine learning will become mainstream are too pessimistic. It’ll happen even faster. And that’s the way it works with exponential trends: they’re slower than we expect, then they catch us off guard and soar ahead.

Andrew McAfee: There’s a passage from a Hemingway novel about a man going broke in two ways: “gradually and then suddenly.” And that characterizes the progress of digital technologies. It was really slow and gradual and then, boom—suddenly, it’s right now.

Jeremy Howard: The difference here is each thing builds on each other thing. The data and the computational capability are increasing exponentially, and the more data you give these deep-learning networks and the more computational capability you give them, the better the result becomes because the results of previous machine-learning exercises can be fed back into the algorithms. That means each layer becomes a foundation for the next layer of machine learning, and the whole thing scales in a multiplicative way every year. There’s no reason to believe that has a limit.

Erik Brynjolfsson: With the foundational layers we now have in place, you can take a prior innovation and augment it to create something new. This is very different from the common idea that innovations get used up like low-hanging fruit. Now each innovation actually adds to our stock of building blocks and allows us to do new things.

One of my students, for example, built an app on Facebook. It took him about three weeks to build, and within a few months the app had reached 1.3 million users. He was able to do that with no particularly special skills and no company infrastructure, because he was building it on top of an existing platform, Facebook, which of course is built on the web, which is built on the Internet. Each of the prior innovations provided building blocks for new innovations. I think it’s no accident that so many of today’s innovators are younger than innovators were a generation ago; it’s so much easier to build on things that are preexisting.

Jeremy Howard: I think people are massively underestimating the impact, on both their organizations and on society, of the combination of data plus modern analytical techniques. The reason for that is very clear: these techniques are growing exponentially in capability, and the human brain just can’t conceive of that.

There is no organization that shouldn’t be thinking about leveraging these approaches, because either you do—in which case you’ll probably surpass the competition—or somebody else will. And by the time the competition has learned to leverage data really effectively, it’s probably going to be too late for you to try to catch up. Your competitors will be on the exponential path, and you’ll still be on that linear path.

Let me give you an example. Google announced last month that it had just completed mapping the exact location of every business, every household, and every street number in the entirety of France. You’d think it would have needed to send a team of 100 people out to each suburb and district to go around with a GPS and that the whole thing would take maybe a year, right? In fact, it took Google one hour.

Now, how did the company do that? Rather than programming a computer yourself to do something, with machine learning you give it some examples and it kind of figures out the rest. So Google took its street-view database—hundreds of millions of images—and had somebody manually go through a few hundred and circle the street numbers in them. Then Google fed that to a machine-learning algorithm and said, “You figure out what’s unique about those circled things, find them in the other 100 million images, and then read the numbers that you find.” That’s what took one hour. So when you switch from a traditional to a machine-learning way of doing things, you increase productivity and scalability by so many orders of magnitude that the nature of the challenges your organization faces totally changes.

The senior-executive role

How will top managers go about their day-to-day jobs?

Andrew McAfee: The First Machine Age really led to the art and science and practice of management—to management as a discipline. As we expanded these big organizations, factories, and railways, we had to create organizations to oversee that very complicated infrastructure. We had to invent what management was.

In the Second Machine Age, there are going to be equally big changes to the art of running an organization.

I can’t think of a corner of the business world (or a discipline within it) that is immune to the astonishing technological progress we’re seeing. That clearly includes being at the top of a large global enterprise.

I don’t think this means that everything those leaders do right now becomes irrelevant. I’ve still never seen a piece of technology that could negotiate effectively. Or motivate and lead a team. Or figure out what’s going on in a rich social situation or what motivates people and how you get them to move in the direction you want.

These are human abilities. They’re going to stick around. But if the people currently running large enterprises think there’s nothing about the technology revolution that’s going to affect them, I think they would be naïve.

So the role of a senior manager in a deeply data-driven world is going to shift. I think the job is going to be to figure out, “Where do I actually add value and where should I get out of the way and go where the data take me?” That’s going to mean a very deep rethinking of the idea of the managerial “gut,” or intuition.

It’s striking how little data you need before you would want to switch over and start being data driven instead of intuition driven. Right now, there are a lot of leaders of organizations who say, “Of course I’m data driven. I take the data and I use that as an input to my final decision-making process.” But there’s a lot of research showing that, in general, this leads to a worse outcome than if you rely purely on the data. Now, there are a ton of wrinkles here. But on average, if you second-guess what the data tell you, you tend to have worse results. And it’s very painful—especially for experienced, successful people—to walk away quickly from the idea that there’s something inherently magical or unsurpassable about our particular intuition.

Jeremy Howard: Top executives get where they are because they are really, really good at what they do. And these executives trust the people around them because they are also good at what they do and because of their domain expertise. Unfortunately, this now saddles executives with a real difficulty, which is how to become data driven when your entire culture is built, by definition, on domain expertise. Everybody who is a domain expert, everybody who is running an organization or serves on a senior-executive team, really believes in their capability and for good reason—it got them there. But in a sense, you are suffering from survivor bias, right?

You got there because you’re successful, and you’re successful because you got there. You are going to underestimate, fundamentally, the importance of data. The only way to understand data is to look at these data-driven companies like Facebook and Netflix and Amazon and Google and say, “OK, you know, I can see that’s a different way of running an organization.” It is certainly not the case that domain expertise is suddenly redundant. But data expertise is at least as important and will become exponentially more important. So this is the trick. Data will tell you what’s really going on, whereas domain expertise will always bias you toward the status quo, and that makes it very hard to keep up with these disruptions.

Erik Brynjolfsson: Pablo Picasso once made a great observation. He said, “Computers are useless. They can only give you answers.” I think he was half right. It’s true they give you answers—but that’s not useless; that has some value. What he was stressing was the importance of being able to ask the right questions, and that skill is going to be very important going forward and will require not just technical skills but also some domain knowledge of what your customers are demanding, even if they don’t know it. This combination of technical skills and domain knowledge is the sweet spot going forward.

Anthony Goldbloom: Two pieces are required to be able to do a really good job in solving a machine-learning problem. The first is somebody who knows what problem to solve and can identify the data sets that might be useful in solving it. Once you get to that point, the best thing you can possibly do is to get rid of the domain expert who comes with preconceptions about what are the interesting correlations or relationships in the data and to bring in somebody who’s really good at drawing signals out of data.

The oil-and-gas industry, for instance, has incredibly rich data sources. As they’re drilling, a lot of their drill bits have sensors that follow the drill bit. And somewhere between every 2 and 15 inches, they’re collecting data on the rock that the drill bit is passing through. They also have seismic data, where they shoot sound waves down into the rock and, based on the time it takes for those sound waves to be captured by a recorder, they can get a sense for what’s under the earth. Now these are incredibly rich and complex data sets and, at the moment, they’ve been mostly manually interpreted. And when you manually interpret what comes off a sensor on a drill bit or a seismic survey, you miss a lot of the richness that a machine-learning algorithm can pick up.

Andrew McAfee: The better you get at doing lots of iterations and lots of experimentation—each perhaps pretty small, each perhaps pretty low-risk and incremental—the more it all adds up over time. But the pilot programs in big enterprises seem to be very precisely engineered never to fail—and to demonstrate the brilliance of the person who had the idea in the first place.

That makes for very shaky edifices, even though they’re designed to not fall apart. By contrast, when you look at what truly innovative companies are doing, they’re asking, “How do I falsify my hypothesis? How do I bang on this idea really hard and actually see if it’s any good?” When you look at a lot of the brilliant web companies, they do hundreds or thousands of experiments a day. It’s easy because they’ve got this test platform called the website. And they can do subtle changes and watch them add up over time.

So one of the implications of the manifested brilliance of the crowd applies to that ancient head-scratcher in economics: what the boundary of the firm should be. What should I be doing myself versus what should I be outsourcing? And, now, what should I be crowdsourcing?

Implications for talent and hiring

It’s important to make sure that the organization has the right skills.

Jeremy Howard: Here’s how Google does HR. It has a unit called the human performance analytics group, which takes data about the performance of all of its employees and what interview questions were they asked, where was their office, how was that part of the organization’s structure, and so forth. Then it runs data analytics to figure out what interview methods work best and what career paths are the most successful.

Anthony Goldbloom: One huge limitation that we see with traditional Fortune 500 companies—and maybe this seems like a facile example, but I think it’s more profound than it seems at first glance—is that they have very rigid pay scales.

And they’re competing with Google, which is willing to pay $5 million a year to somebody who’s really great at building algorithms. The more rigid pay scales at traditional companies don’t allow them to do that, and that’s irrational because the return on investment on a $5 million, incredibly capable data scientist is huge. The traditional Fortune 500 companies are always saying they can’t hire anyone. Well, one reason is they’re not willing to pay what a great data scientist can be paid elsewhere. Not that it’s just about money; the best data scientists are also motivated by interesting problems and, probably most important, by the idea of working with other brilliant people.

Machine learning and computers aren’t terribly good at creative thinking, so the idea that the rewards of most jobs and people will be based on their ability to think creatively is probably right.

About the author

This edited roundtable is adapted from interviews conducted by Rik Kirkland, senior managing editor of McKinsey Publishing, who is based in McKinsey’s New York office.

complex adaptive systems, data saving lives, facts & data points, health market quality, healthcare, policy, rapid learning health systems

The “pay less, get more” era of health care

Excellent summary of current US funding situation…

http://www.vox.com/2014/9/10/6121631/the-pay-less-get-more-era-of-health-care

The “pay less, get more” era of health care

Updated by on September 10, 2014, 1:00 p.m. ET

Health care spending has, for decades, followed a consistent pattern. America pays more and more for health care — and gets less and less.

Between 1990 and 2012, the insured rate in the United States fell two percentage points, from 86.6 to 84.6 percent. If the insured rate had just held steady, six million more people would have been covered in 2012.

While we were covering less people, we kept spending more on health care. National health spending, over that time period, rose from 12 percent of the economy in 1990 to 17.2 percent in 2012. Adjusted for inflation, health-care spending rose from $1.1 trillion to $2.8 trillion over those 22 years.

health spend more get less

That’s been the typical story of American health care: a lousy deal where we get less and spend more.

But there’s a growing body of evidence that this trend is changing; that we’re starting to get a shockingly better deal in a way that has giant consequences for how America spends money. Call it the “get more, pay less” era.

The “get more, pay less” era of health care spending

There are two big trends that, taken together, suggest we may be fundamentally different era of health care spending.

The first is lots more people getting coverage. This is mostly Obamacare: the health care law is expected to expand insurance coverage to 26 million people by 2024. In 2014 alone, most estimates suggest about 5 million people have gained health coverage through the law. The recovering economy is likely playing a supporting role, too, with those gaining jobs also gaining access to employer-sponsored coverage.

The second big trend is in what we spend: actuaries expect that health care costs will grow slower over the next decade than they did in the 1990s and 2000s.

More specifically: health care costs grew, on average, 2 percent faster than the economy between 1990 and 2008. Health spending took over an ever-growing share of the economy. Workers barely got raises; skyrocketing premiums ate up most of their additional wages.

The next decade is now expected to be different. Actuaries at the Center for Medicare and Medicaid Services project health care costs to grow 1 percent faster than the rest of the economy between 2013 and 2023.

“We are seeing historic moderation in costs now over a considerable period of time,” Kaiser Family Foundation president Drew Altman says. HIs group recently released data showing slow growth of employer-sponsored coverage. “It’s absolutely true we’re seeing that and any expert will tell you that.”

This is startling: over the next decade, forecasters think our health spending will grow at a slower rate, even as millions and millions of Americans gain access to health insurance. After two decades of spending more and getting less, we’re entering an era of spending less and getting more. It’s bizarro health spending world.

There are signs of this throughout the health care system

One thing that’s so striking about the “get more, pay less” trend is that it isn’t limited to one particular insurance plan or program. It’s starting to crop up in lots of new health care data, suggesting this change has become pervasive in the health care industry.

Start with private health insurance: the Kaiser Family Foundation recently published research finding the average price of Obamacare’s benchmark will fall slightly in 2015. As my colleague Ezra Klein wrote recently, this just about unprecedented. “Falling is not a word that people associate with health-insurance premiums,” he writes .”They tend to rise as regularly as the morning sun.”

Lower premiums make health care dollars stretch further: Obamacare shoppers will be able to buy the coverage they had last year at a slightly lower price. That’s a big deal when you’re talking about paying for a health insurance program meant to cover tens of millions of Americans.

Increasingly narrow health insurance networks are another sign of “get more, pay less” era. Over the past few years — and especially under Obamacare —insurers have gravitated towards cheaper premium plans to offer access to a smaller number of doctors.

narrow network graph

These plans’ more limited doctor choice can have a big impact on spending. Research from economists Jon Gruber and Robin McKnight found that, in one example, switching enrollees to these plans cut overall spending by one third. And while patients had access to fewer hospitals, the hospitals that were in network were of equally good quality.

Then there’s the Medicare side of the equation, where there has been a unprecedented decline in per person spending. Margot Sanger-Katz at the Upshot has had two fantastic posts on Medicare’s cost slowdown. One of them points out the fact that, since 2010, per patient spending has grown slower than the rest of the economy. You can see that in this graph, which charts “excess cost growth” in Medicare (health wonk speak for cost growth above and beyond inflation). For the past few years, excess growth has been replaced by slower-than-the-economy growth.

medicare excess cost growth

(The New York Times)

As Sanger-Katz points out, there are two trends at play in Medicare. One is that younger baby boomers keep aging onto the program. They’re younger than Medicare’s really old patients, and typically less expensive to care for. That drives down per person spending for the whole population.

But there’s something else going on that looks to be a more permanent trend: Medicare patients are using less expensive care. They go to the doctor more, and the hospital less. You can see this in new data from the Medicare Trustees’ report, which shows per person spending on Medicare Part A (the program that covers inpatient care) falling over the past few years.

medicare

Because of this shift away from hospital care, Medicare Part A now spends less money to cover more people. It paid $266.8 billion covering 50.3 million people in 2012. In 2013, the the same program spent $266.2 billion to cover 51.9 million people.

Will “pay less, get more” health care stick?

We have had periods of relatively slow health care growth before. In the mid-1990s, for example, there was a stretch of time when health spending grew at the same rate as the rest of the economy. You can see that in this graph.

health spending growth

Most health economists attribute that to the rise of health maintenance organizations, or HMOs, that sharply limited access to specialists. Patients, unsurprisingly, didn’t like those limitations and there was a backlash. HMOs declined and health spending rose again.

But some health economists say that this time feels different. For one, the changes are happening in private insurance and Medicare, suggesting there’s no single — and thus easily reversible — force driving the change.

And while there are more patients in narrow network products, something akin to HMOs, consumers are often choosing to be there. These are shoppers on the Obamacare exchanges who have decided to make a trade off: they’re take lower premiums for less choice of doctor.

“In the 1990s, people were essentially stuck in HMOs,” M.I.T economist Gruber says. “This time, people are given an option and make a choice. That’s why I’m more confident this slower growth will stick.”

Medicare actuaries are not fortune tellers; they do not have a crystal ball that conjures up the future of health care with perfect clarity. But at least at this particular moment, there are lots of signs cropping up to suggest something very important in health care is changing, and it’s for the better.

CARD 3 OF 15LAUNCH CARDS

How does American health-care spending compare to other countries?

The United States has higher per-person health-care spending than all other industrialized nations. The most recent international data from the OECD estimates that the United States puts 17.7 percent of its economy towards health care (slightly higher than CMS’s estimate of 17.2 percent). The OECD average is 9.3 percent.

Health_care_oecd

Much of the difference between health care spending abroad and in the United States has to do with prices. Americans don’t actually go to the doctor a lot more than people in other countries. But when we do, our medical care costs more. Specific services, like MRIs and knee replacements, have significantly higher price tags when delivered in the United States than elsewhere.

Health-care spending15 CARDS / EDITED BY SARAH KLIFF UPDATED APR 30 2014, 9:43A
cool, data saving lives

Probabilities of failing birth control methods

 

Probabilities of failing birth control methods

Probabilities of failing birth control methods

SEPTEMBER 15, 2014  |  STATISTICAL VISUALIZATION

Birth control effectiveness

In high school health class, where I learned about contraceptives and the dangers of pre-marital sex, my teacher spouted rates to scare. He would say something like condoms are 98 percent effective but never explained what that meant. Do they break 2 percent of the time? Do couples get pregnant 2 percent of the time? STDs?

These charts from Gregor Aisch and Bill Marsh might help. They show the probability of an unplanned pregnancy, categorized by contraceptive and over a span of ten years. The top solid lines represent probabilities with “typical use” and the dashed lines on the bottom represent probabilities with “perfect use.”

Maybe it’s time for better instructions on how to use these things.

Update: The calculation of long-term probabilities is likely on the pessimistic side and makes too many assumptions about the data and population. Andrew Whitby critiques.

data saving lives, entrepreneurship, health market quality, healthcare, healthy habits, nutrition, policy, quantified self, rapid learning health systems

Bloomberg: Big Data Knows You’ve Got Diabetes Before You Do

 

http://www.bloomberg.com/news/2014-09-11/how-big-data-peers-inside-your-medicine-chest.html

Did You Know You Had Diabetes? It’s All Over the Internet

By Shannon Pettypiece and Jordan Robertson Sep 12, 2014 6:07 AM ET

Photographer: Rick McFarland/Bloomberg

The headquarters of Acxiom Corp. in Little Rock, Arkansas. The Acxiom list was compiled by various sources, including… Read More

Photographer: Joshua Roberts/Bloomberg

An electronic medical records system.

Photographer: Joe Raedle/Getty Images

An elderly man reached for medication in Florida.

Photographer: Joe Raedle/Getty Images

An elderly woman with her medication in Maine.

The 42-year-old information technology worker’s name recently showed up in a database of millions of people with “diabetes interest” sold by Acxiom Corp. (ACXM), one of the world’s biggest data brokers. One buyer, data reseller Exact Data, posted Abate’s name and address online, along with 100 others, under the header Sample Diabetes Mailing List. It’s just one of hundreds of medical databases up for sale to marketers.

In a year when former National Security Agency contractor Edward Snowden’s revelations about the collection of U.S. phone data have sparked privacy fears, data miners have been quietly using their tools to peek into America’s medicine cabinets. Tapping social media, health-related phone apps and medical websites, data aggregators are scooping up bits and pieces of tens of millions of Americans’ medical histories. Even a purchase at the pharmacy can land a shopper on a health list.

“People would be shocked if they knew they were on some of these lists,” said Pam Dixon, president of the non-profit advocacy group World Privacy Forum, who has testified before Congress on the data broker industry. “Yet millions are.”

They’re showing up in directories with names like “Suffering Seniors” or “Aching and Ailing,” according to a Bloomberg review of this little-known corner of the data mining industry. Other lists are categorized by diagnosis, including groupings of 2.3 million cancer patients, 14 million depression sufferers and 600,000 homes where a child or other member of the household has autism or attention deficit disorder.

The lists typically sell for about 15 cents per name and can be broken down into sub-categories, like ethnicity, income level and geography for a few pennies more.

Diaper Coupons

Some consumers may benefit, like those who find out about a new drug or service that could improve their health. And Americans are already used to being sliced and diced along demographic lines. Lawn-care ads for new homeowners and diaper coupons for expecting moms are as predictable as the arrival of the AARP magazine on the doorsteps of the just-turned 50 set. Yet collecting massive quantities of intimate health data is new territory and many privacy experts say it has gone too far.

“It is outrageous and unfair to consumers that companies profiting off the collection and sale of individuals’ health information operate behind a veil of secrecy,” said U.S. Senator Jay Rockefeller, a West Virginia Democrat. “Consumers deserve to know who is profiting.”

Senators’ Attention

Rockefeller and U.S. Senator Edward Markey, a Democrat from Massachusetts, introducedlegislation in February that would allow consumers to see what information has been collected on them and make it easier to opt out of being included on such lists. In May, the Federal Trade Commission recommended Congress put more protections around the collection of health and other sensitive information to ensure consumers know how the details they are sharing are going to be used.

The companies selling the data say it’s secure and contains only information from consumers who want it shared with marketers so they can learn more about their condition. The data broker trade group, the Direct Marketing Association, said it has its own set of mandatory guidelines to ensure the data is ethically collected and used. It also has a website to allow consumers to opt out of receiving marketing material.

“We have very strong self regulation, we have for more than 40 years,” said Rachel Nyswander Thomas, vice president for government affairs for the DMA. “Regardless of how the practices are evolving, the self-regulation is as strong as ever.”

Yet the ease with which data is discoverable in a simple Google search along with Bloomberg interviews with people who showed up in one such database suggest the process isn’t always secure or transparent.

Open Access

Dan Abate said he never agreed to be included in any list related to diabetes. Two other people on the same mailing list said they didn’t have diabetes either and weren’t aware of consenting to offer their information.

In Abate’s case, neither he nor anyone in his family or household has diabetes and the only connection he can think of for landing on the list are a few cycling events he participated in for a group that raises money for the disease.

“I could understand if I was voluntarily putting this medical information out there,” Abate said. “But I don’t have diabetes, and I don’t want my information out there to be sold.”

Bloomberg found the diabetes mailing list on the website of Exact Data in a section for sample lists that included dozens of other categories, like gamblers and pregnant women. The diabetes list contained 100 names, addresses and e-mails. Bloomberg sent e-mails to all of them, and three consented to interviews. There were no restrictions on who could access the list, available on search engines like Google.

Online Surveys

Exact Data’s Chief Executive Officer Larry Organ said the list posted on its website shouldn’t have included last names and street addresses, and the company has since deleted any identifiable information. He said the data came from Acxiom and Exact Data was reselling it.

The Acxiom list was compiled by various sources, including surveys, registrations, or summaries of retail purchases that indicated someone in the household has an interest in diabetes, said Ines Gutzmer, a spokeswoman for the Little Rock, Arkansas-based company. While Gutzmer said consumers can visit the Acxiom website to see some of the information that has been collected on them, she declined to comment about how any one individual was placed on the list.

Acxiom shares rose less than 1 percent, to $18.66 at the close of New York trading. The company has lost 29 percent of its value in the past 12 months.

Sharing Information

One of the more common ways to end up on a health list is by sharing health information on a mail or online survey, according to interviews with data brokers and the review of dozens of health-related lists. In some cases the surveys are tied to discounts or sweepstakes. Others are sent by a company seeking customer feedback after a purchase. The information is then sold to data brokers who repackage and resell it.

Epsilon, which has data on 54 million households based on information gathered from its Shopper’s Voice survey, has lists containing information on 447,000 households in which someone has Alzheimer’s, 146,000 with Parkinson’s disease, and 41,000 with Lou Gehrig’s disease. The Irving, Texas-based company provides survey respondents with coupons and a chance to win $10,000 in exchange for information on their household’s spending habits and health.

The company will share with individual consumers specific information it has gathered, said Jeanette Fitzgerald, Epsilon’s chief privacy officer.

Suffering Seniors

KBM Group, one of the largest collectors of consumer health data based in Richardson, Texas, has health information on at least 82 million consumers categorized by more than 100 medical conditions obtained from surveys conducted by third-party contractors. The company declined to provide an example of the surveys. KBM uses the information for its own marketing clients, and sells it to other data brokers, said Gary Laben, chief executive officer of KBM.

“None of our clients wants to engage with consumers or businesses who don’t want to engage with them,” he said. “Our business is about creating mutual value and if there is none, the process doesn’t work.”

Data repackaging is extensive and pervasive. The Suffering Seniors Mailing List help marketers push everything from lawn care to financial products. It consists of the names, addresses, and health information of 4.7 million “suffering seniors,” according to promotional material for the list. Beach List Direct Inc. sells the information for 15 cents a name. Marketed as “the perfect list for mailers targeting the ailing elderly,” it contains a breakdown of those with diseases like depression, cancer and Alzheimer’s, according to its seller’s website.

Clay Beach, the contact on Beach List’s website, did not return calls and e-mails over the past month.

‘Confidential’ Clients

Little is known about who buys medical lists since data brokers say their clients are confidential, Rockefeller said at a hearing on the issue in December.

Promotional material for the Suffering Seniors data found by Bloomberg on Beach List’s website initially included a list of users. The names of those users have since been removed.

One customer was magazine publisher Meredith Corp. (MDP), which used the list in a test for a subscription offer for Diabetic Living magazine, said Jenny McCoy, a spokeswoman. Other users have included the American Diabetes Association, which said a small portion of names from the list was given to one of its local chapters, and Remedy Health Media, a publisher of medical websites.

Magazine Advertising

Remedy Health may have used the list to advertise one of its magazines, which has been defunct for several years, said David Lee, the company’s executive vice president of publishing.

A growing source of data fodder are website registration forms that ask for health information in order for a user to access the site or receive an e-mail newsletter.

One such site is Primehealthsolutions.com, which provides basic health information on a variety of conditions. It makes money by collecting data on diseases its users have been diagnosed with and medications they are taking, which people disclose when signing up for the site’s e-mail newsletter.

The site has more than three dozen lists for sale, including a tally of 2.2 million people with depression, 267,000 with Alzheimer’s, 553,000 with impotence, and 2.1 million women going through menopause.

Jason Rines, a co-owner of Prime Health Solutions, said he will share the lists only with those marketing health-related products, like pharmaceutical or medical device makers.

Purchasing Trail

Acxiom said it uses retail purchase history or magazine subscriptions to make assessments about whether someone has a particular disease interest.

Health data collection is troubling to people like Rebecca Price, who has early-stage Alzheimer’s disease. While she now makes no secret of her disease and serves as a member of the Alzheimer’s Association’s early stage advisory group, that wasn’t always the case. Price, a 62-year-old former doctor, said she initially didn’t even tell her husband of her condition for fear word would get out and harm her personally and financially.

“It is a very, very personal diagnosis,” Price said.

Social media is another potential way information can be collected on patients, said Dixon, of the World Privacy Forum, who warns patients to be more careful about what they share on sites like Facebook.

“Don’t ‘like’ the hospital website or comment ‘thank you for the great breast cancer screening you gave me,’” she said. “Under the Facebook policy that is public information and it is in the wild and if someone goes to that site and pulls it off, it is totally public.”

Facebook Policy

While it would be possible for data miners to scrape ‘likes’ and public comments from Facebook Inc. (FB)’s social network, the company said such practice is against company policy and, if discovered, would be blocked.

“We don’t allow third-party data providers to scrape or collect information without our permission,” said Facebook spokeswoman Elisabeth Diana. “Third-party data providers that work with Facebook don’t collect personally identifiable information and are subject to our policies.”

For consumers who want to know what list they may be on, there are limited options. KBM for example doesn’t have the technological capabilities to look up an individual by name and tell them what lists they are on, though they can purge a name from all their lists if requested to do so, said CEO Laben.

Acxiom started a website last year that allows people to view some of the information it has on them. Those who choose to can correct or remove their data.

Epsilon’s Fitzgerald says the best way for consumers to protect themselves is to be more aware of where they are sharing their information and pay more attention to website privacy policies.

“If people are concerned, don’t put the information out there,” Fitzgerald said. “Consumers would be better served if they were educated more on what is going on on the web.”

(A previous version of the story mistated the name of the Direct Marketing Association and corrected the spelling of Facebook spokeswoman Elisabeth Diana.)

To contact the reporters on this story: Shannon Pettypiece in New York atspettypiece@bloomberg.net; Jordan Robertson in San Francisco atjrobertson40@bloomberg.net

To contact the editors responsible for this story: Rick Schine at eschine@bloomberg.net Drew Armstrong

complex adaptive systems, data saving lives

High risk of melanoma for airline crew

High risk of melanoma for airline crew
A SYSTEMATIC review and meta-analysis involving more than 250 000 people has found that pilots and air crew have twice the incidence of melanoma compared with the general population. The review, published in JAMA Dermatology, of 19 studies published between 1990 and 2013 reporting data from 1943 to 2008, included more than 266 431 participants from 11 countries. Fifteen of the papers reported data on pilots and four on cabin crew. The researchers found the standardised incidence ratio of participants in any flight-based occupation was 2.21 — 2.22 for pilots and 2.09 for cabin crew. The standardised mortality ratio of participants in any flight-based occupation was 1.42 — 1.83 for pilots and 0.90 for cabin crew. The researchers speculated that cosmic radiation could be a risk factor, saying “UV radiation is a known risk factor for melanoma, and the cumulative exposure of pilots and cabin crew compared with the general population has not been assessed”. They wrote that their findings had “important implications for occupational health and protection of this population”.

https://www.mja.com.au/insight/2014/33/news-brief