Holes in the Sidewalk of Analytics

Analytics needs to walk around some holes in the sidewalk. 

A wonderful book of poems by Portia Nelson, There's a Hole in My Sidewalk: The Romance of Self-Discovery, addresses the struggle to stop falling into the same psychological/behavioral hole, how to walk around it and “go down another street”…and grow as a person.   

The field of analytics has fallen into a few big holes lately that represent both its promise and its peril.  These holes pertain to privacy, policy, and predictions.  

Privacy.  $1B. Target, the retailer, was the poster child for using big data for customer analytics to pump up sales.  It unabashedly collected lots of data on its customers, from a variety of sources, integrated it, and used it for predictive modeling to identify segments that are experiencing “moments that matter” when habits can be influenced to buy new products.  Target touts that “we’ll be sending you coupons for things you want before you even know you want them.”  For example, it developed algorithms about the probability of pregnancy and the delivery date to sell specific products that women buy at different times during their pregnancy.  It identified the women, sent them coupons, and opened its cash registers to amazing profits.  However, as we have learned, it also opened its cash registers, credit card machines, and databases to cybercriminals who stole the personal data of tens of millions of customers.  It is estimated that this error will cost Target over $1B in fraud claims.  Its stock price has fallen over 25% since the incident. 

  

The “hole” is a comfortable one for analytics.   The habit is to uncork technology before its time.   For example, the NSA exploited the technology to tap telephone calls and scrape peoples’ metadata into a database before it confronted the likelihood that world leaders and the public at large would condemn it and it could not defend it in terms of averting terrorism.  Similarly, there was a lot of talk about the “creepiness” of retailers collecting personal data on customers by whatever means possible.  The big appetite for the data to improve sales may have blinded companies from thinking about the consequences and “forgetting” the basic responsibility to protect it.  In the Target case, there are known credit card technology safeguards, including the use of a security microchip, that were ignored.  Additionally, there must be encryption protocols and firewalls to decouple data so that cybercriminals would not find personal identity information.  The simple lesson is that just because the technology exists does mean that it should be used.  Perhaps one route around the “hole” is to “count to ten” before technology genies are let out of the bottle.

Predictions:  43-8.  The great hope to demonstrate the value of analytics is (advanced) predictions.   It uses all the breadth and depth of big data to go beyond reporting on the past to predicting the future.  So, how could the predictions about the 2014 Super Bowl game between the Sea Hawks and the Broncos be so far off?   The point spread was 3 points but the actual spread was more than 10 times that as the Sea Hawks routed the Broncos and Peyton Manning from the first (mis) play of the game.   Perhaps there is a tribe of analytics “sharps” who are making it big in sports wagering but the facts are that the best of them only win about 53% of the time. 

The irony perhaps is that football, like baseball and basketball, is a fully digitized industry unlike most others including healthcare which still struggles to use electronic medical records to capture its key transactions information.  In sports, every play action on the field is captured, recorded, and discussed, resulting in a rich performance database of players in almost every conceivable context, e.g. how a baseball hitter performs relative to a specific pitcher, playing field, regular or post-season game, and so forth. 

But, it is clear from the big-miss prediction of the Super Bowl game that some important data that would improve the precision of the model are missing.  The “squares”, who rely on softer data (intuition), think they know this realty of the shortcomings of quant data, although their win rate is no better than that of the sharps.  My personal insight on this is when I was 16 years old I worked as a dog handler at a greyhound racing park.  I took a dog from its pen, to the viewing stand, into the starting gate, and picked it up at the conclusion of the race.   I knew when the dog was nervous, sick, and hyped up.  And I knew when they hit their head going into the gate that they would not recover to win the race.  

The “hole” here is the reliance on the big data that is under the lamppost.  In this case, it is the big sports data, most of which is collected…because it can be… without a model in mind and mostly for its entertainment value.  The big data presumption is that if you build it (the database), the predictions will come.  That ain’t necessarily so, even if one runs zillions of simulations on all the yottabyte of big data.  The data have to be right for the model to work.  In the case of sports, there are lots of (“soft”) untapped personal data such as health, resilience, and response to certain threats (and more) that may be important factors in big game performance.   It’s a real short circuiting of predictive modeling to be carried away with the technologies of the yottabytes while avoiding a full understanding of the phenomena under study.

Policy.  2.2/7.  The biggest analytics project in recent history is the $6 billion federal investment in the health exchanges.  The goals of the health exchanges are to enroll people in the health insurance plans of their choice, determine insurance subsidies for individuals, and inform insurance companies so that they could issue policies and bills.  The project touches on all the requisites of analytics including big data collection, multiple sources, integration, embedded algorithms, real time reporting, and state of the art software and hardware.  As everyone knows, the implementation was a terrible failure.  The CBO’s conservative estimate was that 7 million individuals would enroll in the exchanges.  Only 2.2 million did so by the end of 2013.  (This does not include Medicaid enrollment which had its own projections.)  The big federal vendor, CGI, is being blamed for the mess.  Note that CGI was also the vendor for the Commonwealth of Massachusetts which had the worst performance of all states in meeting enrollment numbers despite its long head start as the Romney reform state and its groundbreaking exchange called the Connector. New analytics vendors, including Accenture and Optum, have been brought in for the rescue.   

Was it really a result of bad software, hardware, and coding?   Was it  that the design to enroll and determine subsidies had “complexity built-in” because of the legislation that cobbled together existing cumbersome systems, e.g. private health insurance systems?  Was it because of the incessant politics of repeal that distracted policy implementation?  Yes, all of the above. 

The big “hole”, in my view, was the lack of communications between the policy makers (the business) and the technology people.  The technologists complained that the business could not make decisions and provide clear guidance.  The business expected the technology companies to know all about the complicated analytics and get the job done, on time.   This ensuing rift where each group did not know how to talk with the other is recognized as a critical failure point.  In fact, those who are stepping into the rescue role have emphasized that there will be management status checks daily “at 9 AM and 5 PM” to bring people together, know the plan, manage the project, stay focused, and solve problems.  Walking around the hole will require a better understanding as to why the business and the technology folks do not communicate well and to recognize that soft people skills can avert hard technical catastrophes.

In summary, these three holes in the sidewalk of analytics are recurrent themes and threats to fulfilling the promise of analytics.  First, the technology cannot zoom ahead of the sociology.  The need for business results cannot err on the side of the creepy use of personal data to increase sales without a full respect of the need to protect privacy and to honor customers.  Second, big data is not the answer if it is not the right data.  The full potential of predictive modeling requires more thinking and less data processing.  And lastly, the big failures in analytics have less to do with bad machines and buggy software and much more to do with people on either side of the business and technology fence just not talking with one another. 

The Flatlining of Healthcare

The business of healthcare is facing a defining moment.  For the first time in decades, the growth in healthcare expenditures continues to be slower than the rate of inflation.  In 2012 it was nearly one full percentage point lower at 3.7%.  And job growth for the industry is nearly flat at 1.4%.  How will the industry respond?  Will it conclude that this a momentary aberration and the best course is business-as-usual and protect its flank to keep the business viable during the maelstrom?  Or will it consider the likely reality that a line has been crossed and business will never be quite the same again?   

The business of healthcare, up until this point, has been reliably good as judged by its profits (as a percent of revenues) at about 7%.  But, its performance in improving health has been abysmal.  It has the poorest health outcomes when compared to peer countries and the worst efficiency of any industry.  The likelihood of getting the right treatment at the right time is just a little bit better than a coin toss.  And its consumer engagement is the worst of any industry.  It is clear that the American way of the business of healthcare is not always aligned with the production of health for people.

The paradox is that there are great opportunities to improve health outcomes and to do so at significantly less cost, thus improving economic efficiency.  But, there is one humungous fly in the ointment.  Most of these innovations will result in a big loss in the billable services which fuel revenues.

 

Many of these innovations are fueled by analytics.  I concentrate on three that hold great promise to transform health and healthcare over the next 5 years.  (See the McKinsey & Company report for more details.)  These are the big ones and there are many others that fit the category.  The top three include:

•  The combination of mobile computing devices, high-speed wireless connectivity, applications, and sensors to communicate, track, and manage all things related to health.
•  Next-generation genomic sequencing technologies, in combination with big data analytics, and technologies with the ability to modify organisms that will achieve personalized medicine customized to a “patient of one”.  
• Complex analyses and problem solving made possible by advanced computing technology, machine learning, and natural user interfaces that will automate all types of knowledge work. 

McKinsey & Company estimate that more than 20% of patients with cancer, heart disease and diabetes could receive more relevant and effective personalized care including life extension of up to two years through computer-aided differential diagnosis, connected health, sensors for remote monitoring, tailored treatments, and better communications within healthcare and to patients.  This is huge!  And, there are numerous examples of small scale, emerging solutions in all of these areas. 

But, back to that fly in the ointment.  Present worldwide annual revenues for the treatment of chronic illnesses are about $15 trillion.  A 10-20% cut would dramatically reverse the fortunes of many of those providing these services.  Diagnostic technologies will reduce the need for extra-exploratory tests.  Automation can drastically reduce the costs of knowledge workers.  And pinpoint treatments will diminish trial-and-error medicine.  

There are many challenges to operationalize these innovations.  These include the suboptimal digitization of the industry and an electronic health record that cannot yet function as an information hub.   There is a need for substantial skills to extract, aggregate, translate, and integrate multiple data sources.  Extensive research is needed to bring genomics to the bedside.  There are worrisome unintended consequences related to privacy and security.  And of course, the payment system must change to reward the production of better outcomes rather than more and more billable services.

But the biggest obstacle is the entrenched way of doing business in the healthcare industry.  As Uwe Reinhardt, the Princeton health policy sage, observes “Given that every dollar of health care spending is someone’s health care income…there must exist a surreptitious political constituency that promotes…waste.”  The American way of producing health is failing.  The standard way of providing healthcare must evolve to embrace inevitable changes to delivery and payment systems, the adoption of technologies, and the partnership with people as co-producers of health. 

When I talk with analytics leaders on the ground in prestigious healthcare organizations across the country, they have little appetite for considering that the best use of their time and talent is to improve health through analytics.  They concentrate on “business intelligence” to enhance revenues and reduce operational costs.  They do what their bosses ask of them.  And there is not a great demand of them to use analytics to dramatically improve healthcare and its outcomes in the transformational way that is possible.  

 

Some of these companies will be the last ones to have and use a BETA videocassette, a film camera, and a paper medical record.  Their strategic myopia will cause them to miss the moment and stumble in their competitive rank.   Others will “take the road less traveled by” and embrace the use of analytics, first and foremost, as a resource and support to improve outcomes.  And that will make all the difference. 

My op-ed on high insurance exchange was published by the Providence Journal and included below in this blog.  As we approach the 50th anniversary of Medicare, I wanted to reflect on its first year of implementation and enrollment of beneficiaries and compare it with the health insurance exchanges.  In summary, Medicare signed up 99% of its beneficiaries, within 9 months of President Johnson signing it into law, and did so at a cost of $45 per beneficiary (inflation adjusted).  A tough act to follow.   The exchanges have a long way to go and it is just the right time to start to consider how to increase the rate of enrollment for the 40 million Americans still without health insurance.  Please see more in my op-ed published in the editorial pages of the Providence Journal below.

Lessons of R.I.’s high exchange costs

January 26, 2014 01:00 AM

Comments

BY DWIGHT MCNEILL

The enrollment numbers for the health insurance exchanges under Obamacare are in, and they do not paint a pretty picture. The Congressional Budget Office’s projections for enrollment were 7 million for the exchanges and 9 million for Medicaid. The actual numbers are considerably lower at 2.1 million for the exchanges and 4.4 million for Medicaid. Additionally, about 3.1 million young adults got coverage through Obamacare’s rule forcing insurers to cover dependents up to age 26.

Part of the shortfall is from the technology fumbles of getting the website up and running. But a large part of it may be because of the baked-in complexity of the reform itself.

In Rhode Island, the HealthSource RI exchange surpassed its very modest goal of insuring 10 percent of the state’s 55,000 uninsured. But other goals did not fare as well.

The cost of the exchange is very high. Given Rhode Island enrollment and costs to date and projected over the next few years (in order to spread infrastructure investments over time), the administrative cost as a percent of the total cost, including insurance premiums, is more than 15 times that of Medicare at 2 percent and three times that of private insurance (at 10-plus percent). Note that in addition to enrollment, Medicare and insurers also pay huge volumes of medical bills.

The goal of a market-based system is to use the power of competition among insurance suppliers to drive better quality at lower cost. Since Blue Cross and Blue Shield of Rhode Island is the only private insurer in the state’s exchange, this major reason for an exchange is forfeited.

How can the exchange costs be reduced? The cost of operating the Rhode Island exchange will shift from the federal government to the state in 2015. The projected yearly operating cost is about $23 million.

Three possible solutions:

•Run the exchange more efficiently. I suspect that the complexity of Obamacare and its reliance on the existing private insurance system necessitates these high costs.

•Since there is only one insurer in the exchange, perhaps it should do the enrollment, as it does for its core business.

•Divert the cost to other (out-of-state) taxpayers by shifting the exchange responsibility to the federal government, as have 23 other states.

These solutions would reduce the cost to state taxpayers and businesses but would not solve the underlying cost drivers.

The results of the experiment to use exchanges to get people insured are accumulating, and it is becoming increasingly obvious that modifications to Obamacare must be considered. The Affordable Care Act, Section 1332, supports “innovation waivers,” starting in 2017, for states to try new ways to achieve the same goals for coverage and comprehensive and affordable benefits. Some states, including Vermont, Hawaii, Oregon, New York, Washington, California, Colorado and Maryland, are viewing a single-payer system.

Medicare is a single-payer system, and is supported by the vast majority (96 percent) of seniors. When it was implemented almost 50 years ago, it signed up 99 percent of those eligible for benefits within nine months of President Johnson’s signing the bill into law.

The enrollment process had simplicity “baked in” because Social Security knew those who were eligible. The cost to enroll them was a mere fraction ($45 per enrollee) of the cost of the exchanges (estimates run from $1,000 nationally to $5,000 in Rhode Island per enrollee).

Additionally, the annual growth rate of Medicare spending per capita is projected by the CBO to be substantially lower than private health insurance spending between 2012 and 2021 (3.6 percent vs. 5 percent). And over the last 50 years, Medicare has transformed health care delivery and finance with reforms such as a prescription drug benefit, hospital diagnosis-related groups, quality measurement and transparency, and much more.

Prior to the implementation of the health insurance exchanges, there were 55 million Americans uninsured. In 2014, over 40 million remain uninsured. And it is very unlikely that most of these people will ever get health insurance.

Given the impasse in Congress, any consideration of policy modifications to improve access for the uninsured in the foreseeable future is unlikely. It is up to the states.

Rhode Island should join other leading states to address innovative ways to provide insurance more effectively and efficiently. It should not defund its exchange because, at the moment, it offers the best route to lift people out of the risk of not having insurance. But, it should set in motion a process to reincarnate the inevitable solution to health insurance, a single-payer system.

Dwight McNeill, of Little Compton, is visiting professor of health policy and population health at Suffolk University.

Person Centered Analytics for Health.

In my previous blog, Who Am I…for Health’s Sake, I suggested that we are possessed by different selves that behave in unique ways as we navigate healthcare and our health future.  These distinct selves include that of consumer, patient, citizen and customer.  Each of the four selves is well intentioned but does not live up to its potential to improve health.  They fragment our attention, limit our power, put their own needs above the rest, and derail us from taking control of our own health destiny.  In order to achieve our optimal health potential, we must be, in the words of cummings, “nobody but ourselves” and fight against the forces all around us to “make you everybody else.”

This blogs outlines a way forward that that informs, supports, and strengthens people to improve their health through analytics.

The emerging reality is that the American way of producing health is failing because of its fixation on health care, its denial that people are the active ingredient for change, and its slow uptake of technologies. The new reality is that prevention is more important than treatment, behavior change is the reliable pathway to improved outcomes, and information technologies are shifting power to people to become the primary agents of change. 

It’s about health, stupid!

There is greater appreciation that the health of Americans, ranked the lowest among wealthy nations on most measures, will not improve by spending more on health care.  Compelling evidence on the determinants of health show that personal behavior is most important in reducing premature mortality.  In fact it is about three times as important as health care.  Breakthroughs in health will happen by attending to what is obvious to prevent chronic illnesses…diet, exercise, weight, smoking and doing what the doctor says…rather than through advances in new research and clinical care.  But what is obvious has not been easy.

The science of behavior change is improving…dramatically

People need to change their behavior to achieve better health, but our track record has not been good.  We are “just human” and do not always do the rational thing, can be lazy, have other priorities, stick with our habits, and want to fit in.  And despite the best intentions of those who care for us, including providers, payers, and policy makers, we have not cracked the code.  Until now.

Behavioral economics is all the rage.  It puts together what we know about social psychology and economics to come up with powerful solutions that are working.  It digs deep into what drives behavior change and intervenes at key points.  For example, it understands that people have biases for maintaining the status quo, for the present rather than the future, and about “loss aversion”.  It knows that we have difficulty evaluating risk because we exaggerate small probabilities, we respond to positive rewards that are frequent and fun and that sometimes play on regret, and we tend to follow through with things if we make a contract to do so.   Marketers know these things and use it in advertising to make us to buy things.  It’s time for people and their advocates to embrace these tools to improve health.

Technologies put people in control

People are making more decisions for themselves rather than relying on experts because there is more information available, translated just for them, and constantly available through devices such as smartphones.  People do their banking, airline reservations, and stock trading on their own, 24/7, and they can do the same in managing their own health.  In the near future they will be aided by passive sensors that will monitor their health and have their own Siri-like advisor formulate their daily health agenda.  People stay engaged, supported, and challenged through social media and depend on the wisdom of their peers for product reviews rather than relying on marketers.  And the expanding availability of information and its democratization provide a personal analytics platform for behavior change that is more people centric, self-managed, and delivered outside of the usual healthcare structures in the living room, over the phone, and at the coffee shop.

Know me and work with me…or get lost

As the integrated self takes more control of behaviors to improve health, it will need support, but of a different kind.  People will expect everything to be customized to their needs.  They will demand accountability for products and services to work.  They will be an active participant in key decisions.  And with the convergent forces of a new priority on health outcomes and a focus on behavior change, along with enabling behavior sciences and information technologies, they will assume a central and responsible role to improve their health future.  

Stay tuned for my forthcoming book, Person Centered Analytics for Health .

Who Am I…for Health’s Sake?

Sybil is a true story about a woman possessed with sixteen different personalities spanning the intensely dramatic Vanessa to the vivacious Marjorie.  The psychiatric term is dissociative identity disorder which is characterized by at least two identities that alternatively control a person’s behavior.  After considerable treatment, Sybil’s different selves were able to reconcile and Sybil combined them into an integrated self that relieved her turmoil and improved her well-being.

In healthcare, we are possessed by different selves that behave in unique ways.  These distinct selves include that of consumer, patient, citizen and customer.  As with Sybil, we need to understand what our separate selves are up to and evolve an integrated self that minimizes distractions, takes control, and focuses on a healthy future.

Our four selves in health:

1. Consumer:  America is a shopping nation and we rely on our consumer self when we buy goods and services.  The term “consumer” comes from economics and is predicated on the theory of choice.  The theory states that when people have choice and information on price and quality they make rational decisions to optimize their welfare and to stimulate competition to improve efficiency.  The theory works well in most industries, like retail, but poorly in healthcare because the requirements for a healthy market are distorted:  a) There is little consumer choice (employers (mostly) pick health insurance plans, plans select networks and doctors, and doctors pick specialists, hospitals, and treatments), b) people pay for things with other people’s money (insurance and government subsidies), and c) information for consumer decision-making is either absent, irrelevant, or difficult to understand. Nevertheless, we persist in our belief that a consumer-driven, market-based approach produces superior results compared to alternatives including a government approach. 

The latest example is the health insurance exchanges.  Its primary goal is get more people insured and to make markets work better by rewarding insurers that satisfy consumer needs better than the competition.  But, there is little choice among insurers in most markets.  The choice is often among products offered by a single insurer.   The information is limited.  Yes, there is information on prices.  But, there is no information on insurers’ performance in improving health and customer experience and whether one’s doctors are in the narrower networks offered. 

Also, the lower cost insurance plans conceal hidden costs in the form of much higher out-of-pocket costs.  The “new normal” deductible is $2500+ for the individual silver benchmark plan. Classic research from RAND shows that people with deductible plans at this high level use doctors and prescription drugs significantly less and do not discriminate on the services they cut, whether effective or wasteful.  This has led people to question outlandish medical charges, which is good, and for some to “take out their own stitches”, which is not.

Peoples’ welfare is improved through the exchanges mostly because insurance is more affordable due to subsidies and not because their actions as consumers are inducing more competition to drive down costs and improve quality.  So, the consumer self turns out to be an ineffective role that causes a good deal of frustration and churn and a distraction from focusing on what matters most. 

2. Patient:   Our patient self emerges when we receive medical care.  It is defined by the discipline of medicine which takes a disease orientation that relies on deep knowledge of the science of diagnosis and treatment to make people well.  The expert role of the physician defines the pact between patients and doctors:  Doctors “know best” and assume an authoritarian role and patients comply with their doctor’s “orders” and “prescriptions” and assume a dependent role.

     

There are two limitations of the patient self.  The first is that the medical model works well when there is a known treatment for a specific diagnosis.  But in many cases, when the outcomes of alternative treatments are equivalent or equivocal, the choice of treatment should have more to do with the wishes and tradeoffs of the person rather than the opinions of the doctor.  This is when patients and doctors need to practice shared decision making.  But the medical model has not relented much to a patient-centered model that truly empowers patients in decision making. 

The other limitation is that the medical model only works well when people are sick.  But the majority of sick care today does not stem from pathogens or mysterious medical causes.  Most is for chronic illnesses that are caused by individual’s behavior where prevention and self-monitoring are more important than treatment.  And the medical model has not been very successful in shaping people’s behaviors.  For example, the probability that people will take their medications is 50/50.  And admonitions to eat well and take off pounds do not go far enough.  So, the patient self needs to evolve dramatically to be more actively involved in co-producing health.

3. Citizen:  Our citizen self is expressed when we vote to have others represent our views in the political process and when we participate directly as a member of a community.   It is based on the discipline of political science and the premise that democracy leads to improvements in the status quo. 

Health care has certainly been on the political agenda for the last few election cycles and is positioned for the next. But the citizen self has been relatively passive and on the receiving end of thunderous propaganda from special interests to garner support for their positions.  A slim majority votes in presidential elections and far fewer are involved at the state and local levels.

Although the citizen self is dormant today, there was a time during the 1960s and 1970s when it was in full flower.  One example was the Oregon Health Plan which included a great deal of citizen deliberation about setting priorities for healthcare including what services would be paid for under Medicaid.  The belief was that the only way to control costs was to understand that resources are limited, trade-offs are needed, and the political process must activate deep citizen participation to succeed.   This movement reached its pinnacle during the Great Society era and died off when market oriented approaches to societal challenges supplanted government approaches in the early 1980s. 

The citizen self has been hollowed out and will be not be resurrected unless the playing field is leveled and those in power invite genuine participation.

4. Customer:  A customer is similar to a consumer in that they both buy goods and services but is different because of the underlying discipline that defines it, business.  Business relates to customers in two distinct ways.  In one way, business reveres the customer and keeps them happy with low prices, high quality, and good service in order to build loyalty and profits.  In this view, the customer is always right and close relationships with them can reveal how to improve products and services and develop new ones that meet demand.  In the words of Mahatma Gandhi, “A customer is the most important visitor on our premises…We are not doing him a favor by serving him. He is doing us a favor by giving us an opportunity to do so.”

In the other way, business uses marketing and advertising tactics to deceive the customer.   They use intrusive ways to gather more information about them, without consent, to know them “intimately” in order to sell more.

In healthcare, people are seldom referred to as customers.  After all, they do not account for much of the buying.  Employers buy from insurers, insurers pay doctors, doctors determine treatments.  In business, leverage comes to those with the most money in play.  People are bit players and more likely to be on the receiving end which limits the opportunities for influence and maximizes the likelihood of manipulation.

Our integrated self

Each of the four selves is well-intentioned but does not live up to its potential to improve health.  They fragment our attention, limit our power, put their own needs above the rest, and derail us from taking control of our own health destiny.  To achieve an integrated self, one must understand and balance competing demands and align these with an overarching conviction to achieve our full human potential.  This is not easy.  As  e.e cummings said, “ To be nobody-but-yourself--in a world which is doing its best, night and day, to make you everybody else--means to fight the hardest battle which any human being can fight; and never stop fighting.” 

The good news is that there are developments in three areas that are converging to make the fight winnable.  In my next blog, I will address these promising solutions.